Antigenicity and immunogenicity of multiple antigen peptides (MAP)containing P. vivax CS epitopes in Aotus monkeys

SOCRATES HERRERA1,3, CECILIA DE PLATA1, J. MARIO GONZALEZ2, BLANCA LILIANA PERLAZA 1,FLORENCE BETTENS4, GIAMPIETRO CORRADIN3 & MYRIAM ARE VALO-HERRERA1

1 Instituto de Inmunologı´a del Valle, Universidad del Valle, AA 25360, Cali, Colombia2 Fundacio´n Centro de Primates, Universidad del Valle, AA 2188 Cali, Colombia3 Institute of Biochemistry, University of Lausanne, Ch. des Boveresses 155, 1066 Epalinges, Switzerland4 Institut fur Klinische Immunologie, Inselspital, Bern, Switzerland

SUMMARY

Using linear synthetic peptides corresponding to thePlasmo-dium vivaxcircumsporozoite (CS) protein of the common type,we have identified several T and B-cell epitopes recognized byhuman individuals. Three T-cell epitopes studied (p6) from theamino, (p11) from the central and (p25) from the carboxylregions, were widely recognized by lymphocytes of immunedonors. A series of six peptides, in addition to p11, representingthe central repeatdomainof theCS(p11-p17) protein wereusedin ELISA assays to map the B-cell epitopes of this region. P11was the peptide most frequently recognized by sera containingantibodies to the homologous CS protein as determined byIFAT. The sequences corresponding to peptides p6, p11 andP25 as well as that representing a universal T-cell epitopederived from the tetanus toxin were used to assemble eightdifferent Multiple Antigen Peptides (MAP). The immunogeni-city of these MAP was analysed inAotusmonkeys. Groups oftwo animals were immunized with each MAP and both antibodyresponse, T-lymphocyte proliferation andin vitro g-IFN pro-duction were evaluated. Two MAPs containing the same B-cellepitope and either a promiscuous CS-protein derived T-cellepitope (p25) or the tetanus toxin epitope (p-tt30) proved to bethe most immunogenic and induced high levels of anti-peptideantibodies that recognized the native protein. Except for ani-mals immunized with MAP VII, there was no correlationbetween antibody levels, lymphocyte proliferation org-IFNproductionin vitro. The broad recognition of these epitopes byindividuals which had been exposed to malaria, the capacity oftheseMAPs to induce antibodies, recognize the cognate protein,and in vitro g-IFN production encourages further analyses ofthe potential of these proteins as malaria vaccine candidates forhuman use.

Keywords malaria vaccine, CS protein, MAP, T cell

INTRODUCTION

Plasmodium vivaxis a malaria parasite species predominantin Latin America, Southeast Asia and Oceania. It causes achronic disease where fatalities are rare but morbidity isnormally high, therefore producing a great economicalimpact in endemic countries.

The sporozoite is the invasive stage of the malariaparasite, and is inoculated into the vertebrate host by thebite of Anophelesmosquitoes. Once in the bloodstream, theparasites are rapidly cleared from circulation where theyinvade liver parenchymal cells, and transform into exo-erythrocytic parasites. By asexual schizogony, the parasitedevelops and bursts the hepatocytes releasing merozoitesthat invade erythrocytes.

Immunization with irradiated malaria sporozoites haveprotected animals (Nussenzweiget al. 1967) and humans(Clyde 1975) against challenge with normal sporozoites.Both cellular and humoral immune responses have beenshown to be necessary for this protective immunity. Transferof monoclonal and polyclonal antibodies directed against thecentral repeat domain of theP. bergheicircumsporozoite (CS)protein have partially or totally protected mice against thechallenge with the homologous parasite (Potocnjaket al.1980, Eganet al. 1987). In human malaria it has beendemonstrated that monoclonal antibodies directed againstthe repeat region of theP. falciparumandP. vivaxneutralizeand abolish their infectivityin vitro (Hollingdaleet al.1987).

Analysis of theP. vivaxCS protein from different regionsof the world have shown that the central repeat domain ischaracterized by two different repeat sequences sharingpartial amino-acid homology. These two sequences denomi-nated common (VK 210) and variant (VK 247) types arepresent in different frequencies in endemic areas (Rosenberget al.1989, Mannet al.1994). In addition, it has been foundthat theP. vivaxCS gene displays multiple single mutationsspread through its complete sequence, however, the immuno-logical relevance of this genetic polymorphism has not yetbeen defined (Mannet al. 1995).

Parasite Immunology, 1997:19: 161–170

q 1997 Blackwell Science Ltd 161

Correspondence: Socrates HerreraReceived: 19 December 1995Accepted for publication: 10 December 1996

Although, antibodies against these different variants, aswell as against other regions of the CS protein (Frankeet al.1992) and other sporozoite proteins (Khusmithet al. 1991)may be important for protection under natural conditions,passive transfer of a monoclonal antibody that specificallyrecognizes the AGDR amino-acid sequence mapping in therepeat domain of the common type of theP. vivaxCS (Arnotet al. 1988) has shown to be protective in Saimiri monkeys(Charoenvitet al. 1991).

Functional immunity in nature may require the simulta-neous co-operation of T-cell components. T-cell depletion ofmice renders the animals completely susceptible to sporozoitechallenge (Weisset al. 1988), whereas complete protectionhas been obtained by passive transfer of CS specific cloned Thelper cells (Del Giudiceet al. 1990) and cytotoxic Tlymphocytes (CTL) (Romeroet al. 1989, Weisset al.1990). Sporozoite specific CTL andg-IFN injection haveshown to prevent the development of parasite inside the livercells (Schofieldet al. 1987, Kumaret al. 1988).

During the last few years, several vaccine trials have beenperformed in humans with both recombinant and syntheticpeptides with the aim of inducing antibody-based protectiveimmunity (Ballou et al. 1987, Herringtonet al. 1987).The protection achieved in these volunteers against theparasite challenge was very poor probably due to the lowlevels of anti-sporozoite antibodies developed.

Recently, the design of Multiple Antigen Peptide (MAP)systems has allowed the incorporation of multiple B andT-cell epitopes in a single immunogenic high molecularweight construct (Tam 1988). MAP containing multiplecopies of a T helper (Th) epitope alone or together withthe immunodominant B epitope of theP. berghei andP. yoelii CS protein induced high titres of anti-peptideantibodies that cross reacted with the cognate antigen andsignificantly protected immunized mice (Tamet al. 1990,Migliorini et al. 1993, Wanget al. 1995).

With the aim of identifying epitopes that could be includedin MAP constructs to be used as vaccine against the humanmalaria parasiteP. vivax, we have studied the antigenicity of aseries of linear peptides and MAP constructs based onP. vivaxCS epitopes alone or in combination with a universal T-cellepitope derived from tetanus toxin (tt). The immunogenicity ofthese MAP constructs was assessed inAotus monkeys andboth antibody production, T-cell proliferation andg-IFNproduction were evaluated.

MATERIALS AND METHODS

Human subjects and blood samples

Human adults (25–50 years old) from the Colombian Pacificcoast, a malaria endemic region, were analysed to determine

both humoral and cellular responses to theP. vivax CSprotein. These individuals were healthy at the time of thestudy but had previously suffered several malarial episodesas recorded in their clinical histories and confirmed serolo-gically. Thirty volunteers who had no history of malariainfection and living in a non- endemic region were includedin the control group. For T-cell proliferation studies, 10 mlof whole blood were obtained from a group of theseindividuals and collected on anticoagulant. Blood wasfractionated into plasma and mononuclear cells on aFicoll-Hypaque density gradient. Plasma samples werestored frozen at¹708C until use, whereas cells wereimmediately used to determinein vitro T-cell proliferation.

Synthesis and characterization of MAP

Nine linear peptides corresponding to sequences mappingeither in the amino and carboxyl flanks, or in the centralrepeat domain of theP. vivaxCS protein were synthesizedusing the F-moc procedure (Athertonet al. 1981). Three ofthese peptide sequences had been previously shown tocorrespond to epitopes recognized by human T-cells(Herreraet al. 1992) and were located in positions 51–70(peptide 6: p6), 96–104 (p11), and 346–365 (p25). Seven18-mer peptides each (p11–p17) corresponding to dimers ofthe different nonamer blocks of the central repeat domain ofthe common type CS protein (Arnotet al. 1991) were usedto map the B-cell epitopes. In addition, a universal T-cellepitope derived from tt (p-tt30) (FNNFTVSFWLRVPK-VSASHLE) (Demotzet al. 1989, Panina-Bordignonet al.1989) was included. Eight tetra-branched MAPs weresynthesized on a lysine core using glycine residues asspacers. MAP I and II are dimers or trimers of the B-cellepitope p11 respectively. MAP III and IV contained mono-mers of T-cell sequences comprised either of p25 or p-tt30,and MAP V was assembled containing the T-cell epitopesp25 and p6. The remaining MAP VI to VIII were assembledusing different combinations of the previous epitopes asfollows: MAP VI contained p11 and p25, MAP VII based onp11 and p-tt30, and MAP VIII contained p11, p25 and p6(Figure 1).

Animals and immunization

SixteenAotus lemurinusmonkeys from Colombian originand bred in the animal facility of the Universidad del Valle(Cali, Colombia) were used. Groups of two monkeys wereimmunized with 100mg of each MAP diluted in PBS andadsorbed to aluminum hydroxide (Merck, Darmstadt, Ger-many). A total volume of 500ml/dose was distributed inmultiple subcutaneous inoculations of the thorax and abdo-men of the animal. Immunizations were done on days 0, 30

S.Herreraet al. Parasite Immunology

q 1997 Blackwell Science Ltd,Parasite Immunology, 19, 161–170162

and 60, and an additional boost was done five month later.Monkeys were bled immediately before each immunizationand 10 days after the last booster dose. Blood obtained byfemoral venipuncture was placed into heparinized tubes andfractionated into plasma and cells to determine antibodylevels, in vitro T-cell proliferation andg-IFN productionrespectively.

Antibody response

The presence of anti-malarial antibodies in both humanand primate sera was determined by an indirect immuno-fluorescence assay (IFAT) using air driedP. vivax sporo-zoites expressing the homologous CS protein as antigen anda goat anti-human IgG-FITC as secondary antibody. Thepresence of natural human antibodies against the centralrepeat domain of the CS protein was determined by ELISAusing as antigens the linear peptides p11-p17 spanning thecentral CS domain, as described below. The antibodyresponse of immunized monkeys was also measured byELISA using the different MAP constructs as antigen.Each sample was tested not only against its correspondingMAP but also against all of the others used in the experiment.Briefly, 96-well microtitre plates were coated with 1mg/ml ofeach MAP diluted in PBS and incubated overnight at 48C.Plates were blocked with 1% BSA in PBS for one h at 258C.After several washes, serial dilutions of test and controlplasma made in PBS-1% BSA-0·05% Tween 20 wereadded and incubated once more for 1 h at 258C. A goat

anti-human Ig alkaline phosphatase conjugated (Sigma,St. Louis MO, USA) was used as second antibody andincubated for one h at 258C. After washing, 100ml ofO-phenylenediamine substrate was added and incubatedfor one h. Absorbance was read at 492 nm in a TitertekMultiskan reader (Flow Laboratories, Mclean, VA, USA).Plasma samples were considered positive when theirabsorbances were above the mean of a pool of normalmonkey plasma plus 3 standard deviations (SD) (Herreraetal. 1994).

Lymphocyte proliferation assay

Peripheral blood mononuclear cells (PBMC) from humanand monkeys were isolated from whole blood samples byFicoll-Hypaque (Sigma St. Louis, MO, USA) densitygradient.Aotus blood samples were taken either prior toimmunization or 10 days after booster doses. Cells wereresuspended at 3×106/ml in RPMI 1640 medium supple-mented with 2 mM L-glutamine, 0.1% non-essential ami-noacid solution, 1 mM sodium pyruvate, 100 U/ml penicillin,100mg/ml streptomycin (Sigma St. Louis, MO, USA) and10% FCS. Each MAP was diluted in the same medium andadded to 100ml of cell suspension in triplicate cultures atfinal concentrations of 100 and 10mg/ml. The individualpeptides included in each MAP were also tested in theseassays at a concentration of 10mg/ml. Cells were cultured in96-well plates (Falcon 3072, Oxnard, CA, USA) at 378C in5% CO2/95% air for four days. Cultures were then labelled

Volume 19, Number 4, April 1997 CSP. vivaxbased Multiple Antigen Peptides

q 1997 Blackwell Science Ltd,Parasite Immunology, 19, 161–170 163

p11

Lys – Ala – OH VIII

Lys – Ala – OH VII

Lys – Ala – OH VI

Lys – Ala – OH V

Lys – Ala – OH IV

Lys – Ala – OH III

Lys – Ala – OH II

Lys – Ala – OH I

GDRADGQPAGDRADGQPA

p6 HVGQSASRGRGENPDDEE

p25 VRRRVNAANKKPEDLTLNDL

ptt-30 FNNFTVSFWKRVPKVSAAHLW

Figure 1 Schematic representation of the tetra-branched MAP constructs. MAP I and II are monoepitopic and contain the B-cell epitope p11;MAP III, IV are monoepitopic and based on T-cell epitopes p25 and p-tt30 respectively; MAP V contains T-cell epitopes p6 and p25; MAPs VI,VII and VIII contain combinations of the B and T-cell epitopes (p6, p11, p25 and p-tt30).

with 1mCi/well of tritiated methylthymidine (New EnglandResearch Products) for 18 h and then harvested onto glassfibre filters strips (Cambridge Technology Inc., Watertown,MA, USA). Filters were dried and processed for liquidscintillation counting. Supernatants of monkey PBMC cul-tures were harvested on day 4 and analysed for the presenceof g-IFN by ELISA. Leucocytes from some of the respondingdonors were typed for HLA-DR haplotypes using the PCRtechnique as described elsewhere (Gaoet al. 1990).

g-IFN assay

A solid-phase sandwich immunoenzymatic assay was usedto determine thein vitro production ofg-IFN by AotusPBMC. Briefly 96-well microplates were coated with amouse anti-humang-IFN monoclonal antibody and incu-bated at 258C for 24 h. After washing, plates were blockedwith 1% BSA into a 200 mM Tris/HCl, pH 7.5 solution.Supernatants of monkey PBMC cultures and serial dilutionsof a humang-IFN standard were added in quadruplicates. Aperoxidase conjugated antibody was added and plates wereincubated at 48C for 24 h. An antibody-peroxidase testsolution containing free anti-g-IFN antibody was added inparallel wells and used as verification test as describedbefore (Gallatiet al. 1987). The enzymatic reaction wasdeveloped with 20 mM of tetramethylbenzidin/H2O2 sub-strate (Sigma, St. Louis MO, USA) and stopped with 1.0 Mof sulphuric acid. Absorbance was read at 450 nm in aTitertek Multiskan reader (Flow Laboratories, McLean,

VA, USA). Results were expressed as the differencebetween the optical density values (OD) obtained in thedirect and verification tests, compared to the curve obtainedfrom theg-IFN standard dilutions.

RESULTS

Identification of human B and T-cell epitopes

We had previously identified several T-cell epitopes recog-nized by a limited number of immune individuals from amalaria endemic area of Colombia (Herreraet al. 1992). Inthe present study we have concentrated the analysis on theepitopes represented by peptides p6, p11 and p25, andexpanded the study to 95 individuals, 65 from an endemicregion and 30 from a non-endemic area. In addition, 20individuals from the same malaria endemic area werestudied to determine theirin vitro response to the universalT-cell epitope ptt-30. All these individuals reported to havebeen immunized in their childhood with the standard DPTvaccine containing TT. As seen in Figure 2, all threepeptides derived from the CS protein were able to stimulatea high percentage of the immune donors (45–60%).Although all three were also capable of inducingproliferation of lymphocytes from some non-immunedonors, only p-25 was widely recognized by these indivi-duals (12/30 donors). HLA-DR typing of a group of 7 ofthese responder donors confirmed a broad range of haplo-types involved in the recognition of such epitopes (Table 1).

S.Herreraet al. Parasite Immunology

q 1997 Blackwell Science Ltd,Parasite Immunology, 19, 161–170164

ptt-30

1008020 6040

Percent of responders

p25

p11

p6

Pep

tid

es

Figure 2 Human T cell response against identifiedP. vivaxCS and TT derived epitopes. Each malarial peptide was tested forin vitro proliferationof PBMC of immune individuals (open bars) (n ¼ 65) and non-immune donors (n ¼ 30) (shadowed bars) donors. All individuals (n ¼ 20) testedwith ptt-30 (black bars) had been previously immunized with TT. Responses equal or above 3.0 SI were considered positive.

PBMC of twenty individuals from the same region werestimulatedin vitro and responded to TT, 17 of them (85%)also responded to the ptt-30.

Human B cell response

Sera from 196 individuals from the endemic area that hadpreviously screened by IFAT for the presence of antibodiesto P. vivaxblood forms, were used to select the donors forthis study. Although all individuals studied had history ofmalaria only 167 (85%) had detectable antibodies againstP.vivax blood stages and 32 of them (19%) presentedantibodies against sporozoites expressing the homologousCS protein type as determined by IFAT. All of these lattersera were analysed by ELISA to determine the titres ofantibodies directed to the peptides representing the centralrepeat domain of the protein. Most of the sera (30/32)

recognizing the common type of CS protein on IFAT alsoreacted with one or another of the 7 peptides. P11 washowever the most commonly recognized peptide (Figure 3).

Efficacy of the MAP constructs to induce antibodyresponse inAotus

Animals immunized with MAP VI and VII produced highanti-peptide antibodies soon after the first immunization,and titres increased with each booster dose. Of the twoanimals immunized with MAP VI (p11+p25),AotusM-25presented a lower antibody titre during the whole experi-ment whereas M-17 displayed a high antibody titre after thethird dose. MAP VII induced higher antibody titres thanMAP VI specific for the immunogen used (MAP VII) andthe native antigen as assessed by the IFAT. All of the otherpeptides were poorly immunogenic and although someof them induced detectable antibodies after the secondimmunization, they remained low even after the fourthimmunization. Except for sera specific for MAP II (3repeat of the B-cell epitope p11) that also recognized thenative antigen, none of the remaining sera reacted withsporozoites by the IFAT test (Table 2).

The specificity test of some of the sera demonstratedrecognition not only to the corresponding MAP but also tothe related MAP. Animals immunized with MAP VI and VIIrecognized MAP II (p11). In the case ofAotusimmunizedwith MAP III, the reaction of these sera with the relatedMAP V was better than against the corresponding peptidein several independent assays. As already seen in mice(Herrera et al. 1994) animals immunized with MAP IVbased on the single T-cell epitope ptt-30 did not producespecific antibodies, however,Aotus immunized with MAP

Volume 19, Number 4, April 1997 CSP. vivaxbased Multiple Antigen Peptides

q 1997 Blackwell Science Ltd,Parasite Immunology, 19, 161–170 165

Table 1 HLA Class II typing and T-cell response of immune donors

Patients Specificity p6 p11 p25

40 DR2 3.9 0.9 1.553 DR2/DR3 1.3 5.1 1.372 DRW11/DRW13 1.4 0.4 3.173 DRW11/DRW7 4.5 8.4 7.874 DR2/DRW13 4.9 1.1 5.176 DR3/DR7 8.4 2.6 7.779 DR1/DR4 1.6 0.8 4.5

105 DR1/DR4 3.2 3.8 3.9114 DR4/DRW14 6.5 2.1 8.9

HLA-DR haplotypes of individuals responding to the CS T-cellepitopes p6, p11 and p25. SI induced by each peptide are shown.Positive results (SI> 3.0) are in bold.

p17 GNGAGGQAA

30

Number of responders

Rep

eat

do

main

10 20

p16 GNGAGGQPA

p15 GDRAAGQAA

p14 GDGAAGQPA

p13 GDRAAGQPA

p11 GDRADGQPA

p12 GDRADGQAA

Figure 3 Mapping of B-cell epitopes on therepeat region. Peptides p11-p17 are dimers ofthe different blocks of the CS central repeatregion. Antibody titres were measured byELISA. Plates were coated with 10mg/ml ofeach peptide and plasma samples tested at 1/100 dilution in 1% BSA-PBS. Samples wereconsidered positive when absorbance at OD492

was above the mean reading of a pool ofnormal human plasma samplesþ 3 SD.

III based on p25 alone produced antibodies against thispeptide. In several cases the two animals immunized withthe same peptide produced significantly different antibodytitres.AotusC-21 but not F-138 immunized with MAP VIIIproduced a weak antibody response that was directed to theepitope p11 as deduced from the reaction with MAP II(Table 3).

T-cell response inAotus

We investigated the proliferative response ofAotusmonkeys to the different MAPs as well as to the individualpeptides. As shown in Table 4, animals immunized withMAP V assembled with two T-cell epitopes induced thehighest stimulation indices, whereas MAP IV, VI and VIIIinduced a poorin vitro response and the remaining ani-mals represented moderate responses. Independent of theintensity and specificity of lymphocyte proliferation, mostanimals presented a significant increase in the proliferativeresponse after immunization, however no consistent pattern ofresponse was obtained in these experiments (Data not shown).

Supernatants of thesein vitro cultures were screenedfor the presence ofg-IFN. Similarly to what was observedwith antibody titres and proliferation assays, the productionof g-IFN in vitro was very irregular and did not correlate

with these other parameters. However, high titres ofg-IFNwere induced in most animals immunized with MAP con-structs containing p25. Although, MAP IV based on ptt-30alone did not induce detectable levels ofg-IFN, PBMC fromanimals immunized with MAP VII produced high levels ofthis cytokine. Aotus immunized with MAP VI and VIIproduced both high antibody titresin vivoandg-IFN in vitro.

DISCUSSION

The analysis of a larger number of donors in this studyallowed us to confirm our previous findings which indicatedthat the epitopes p6, p11, p25 were broadly recognized bylymphocytes of individuals from endemic areas (Herreraet al. 1992). P6 induced strongin vitro proliferationresponses in immune donors and was not predominantlyrecognized by cells from non-immune donors. In contrast,p25 was widely recognized also by individuals that hadnever been in contact with malaria. The recognition of T-cellepitopes by lymphocytes of naive individuals may be impor-tant for a stronger priming of the immune response duringvaccination in such individuals. The response of non-immunedonors to malarial T-cell epitopes has also been seen byseveral groups. (Sinigagliaet al. 1988, Currieret al. 1992,Lopezet al. 1996). Our results also indicated that p11 was not

S.Herreraet al. Parasite Immunology

q 1997 Blackwell Science Ltd,Parasite Immunology, 19, 161–170166

Table 2 Humoral immune response ofAotusmonkeys immunized with MAP

Immunizationa

MonkeyEpitope Immunogen Code First Second Third Fourth IFATb

I A 32 0 0 100 100 0B C 15 0 0 100 100 0

II F 146 0 0 100 1000 10M 1 0 0 100 500 10

III M 147 0 100 1000 5000 0V 05 0 100 100 500 0

T IV M 53 0 0 0 0 0M 57 0 0 0 0 0

V M 65 0 500 100 0 0M 132 0 5000 5000 1000 0

VI M 25 100 100 100 500 10M 17 5000 1000 5000 5000 80

B þ T VII M 27 500 10000 10000 40000 7120M 43 5000 5000 5000 20000 1280

VIII C 21 0 0 0 100 0F 138 0 0 0 0 0

a Antibody titres of each monkey plasma samples measured by ELISA. Plates were coated with 1mg/ml for each MAP. Results are expressed asthe endpoint titre of the highest serum dilution given an absorbance>0.129 (mean of normal sera +3 SD).b IFAT was carried out usingP. vivaxdried sporozoites. The results obtained represent the geometric mean titre of individual monkey plasma afterthe fourth immunization.

Volume 19, Number 4, April 1997 CSP. vivaxbased Multiple Antigen Peptides

q 1997 Blackwell Science Ltd,Parasite Immunology, 19, 161–170 167

Table 3 Specificity of anti-MAP antibodies obtained after 4th immunization

MonkeyEpitope Immunogen Code I II III IV V VI VII VIII

I A 32 0.14 0.01 0.23 0.01 0.01 0.01 0.02 0.07B C 15 0.16 0.02 0.05 0.04 0.02 0.02 0.02 0.01

II F 146 0.15 0.36 0.08 0.06 0.08 0.16 0.17 0.19M 1 0.04 0.32 0.09 0.07 0.09 0.19 0.21 0.05

III M 147 0.11 0.11 0.17 0.05 0.89 0.17 0.05 0.11V 05 0.01 0.07 0.18 0.05 0.55 0.18 0.04 0.08

T IV M 53 0.06 0.04 0.04 0.11 0.11 0.03 0.05 0.04M 57 0.04 0.09 0.03 0.07 0.12 0.06 0.08 0.04

V M 65 0.06 0.09 0.03 0.05 0.14 0.02 0.01 0.06M 132 0.01 0.05 0.01 0.01 1.73 1.16 0.07 1.16

VI M 25 0.06 0.22 0.22 0.01 0.08 0.22 0.07 0.26M 17 0.04 0.87 1.34 0.11 0.11 >2.1 0.42 >2.1

B þ T VII M 27 0.09 1.26 0.09 0.06 0.08 1.13 >2.1 1.72M 43 0.11 0.96 0.08 0.01 0.06 1.26 >2.1 1.33

VIII C 21 0.03 0.14 0.04 0.02 0.01 0.16 0.14 0.25F 138 0.03 0.05 0.03 0.04 0.03 0.03 0.13 0.06

Fine specificity of antibodies inAotusmonkeys sera obtained ten days after the fourth immunization was determined by ELISA. Plates werecoated with the different MAP at 1mg/ml. Plasma from each immunized monkey was diluted 1:100 and analysed against its corresponding MAPand other constructs. The mean of the negative control is 0.07 OD. Cut-off values for positive response is a mean of negative controlþ 3 SD,which corresponds to 0.13 OD.

Table 4 Proliferation and gamma interferon production of MAP immunized monkeys

MAP mg/ml Individual peptide/10mg/mlMonkey g-IFN

Epitope Immunogen Code 100 10 p6 p11 p25 p30 pg/ml

I A 32 4.5 3.4 11.3 400B C 15 4.4 9.1 8.2 150

II F 146 0.7 0.5 9.3 0M 1 2.8 7.1 4.6 0

III M 147 2.3 0.3 1.5 5850V 05 3.6 0.1 1.5 2400

T IV M 53 1.6 1.4 0.4 0M 57 1.5 2.8 0.7 0

V M 65 2.5 90.2 ND 6.5 9350M 132 3.4 3.3 ND 4.8 1050

VI M 25 0.5 0.1 1.3 0.3 560M 17 0.6 3.7 0.7 9.3 7200

B þ T VII M 27 12.4 19.4 18.6 10.2 200M 43 4.4 9.8 11.2 17.7 4400

VIII C 21 0.2 1.1 0.6 0.5 0F 138 1.1 1.2 4.4 0.4 0

T cell proliferation of PBMC from immunized monkeys performed 3 days after the fourth immunization. Results are expressed as the SI of thePBMC from eachAotusmonkey after stimulation with different concentrations (10, 100mg/ml) of each MAP and at 10mg/ml of individualpeptide. Values above 3·0 SI are considered positive.g-IFN was determined by ELISA in supernatants collected from these cultures on day 4.

only a T-cell epitope, but also the most commonly recognizedB-cell epitope of the repeat domain. In addition to thesemalarial epitopes, we could confirm that the TT derived ptt-30epitope was recognized by the majority of the studiedindividuals from the malaria endemic area.

The definition of these broadly recognized epitopesallowed us to use the MAP strategy as well as severalcombinations of such epitopes and tested for their immuno-genicity in animal models. The results of this study are inagreement with a preliminary analysis carried out usingBALB/c mice (Herreraet al.1994). In both studies we havefound that two of the MAP containing combinations of Band T-cell epitopes (MAPs VI, VII) were the most immuno-genic. Both MAPs induced high anti-peptide antibody titresthat were able to recognize the native CS protein on driedsporozoites. MAP VII that contained theP. vivaxCS B-cellepitope (p11) and the TT derived T-cell epitope (ptt-30)induced higher antibody titres than MAP VI, which includedp11 and p25 epitopes both from the CS protein. Althoughthe antibody titres developed byAotusin this study could beconsidered high, differences were observed between ani-mals in response to the same antigen. These differences arelikely to be due to the fact that monkeys are outbred animalsbearing different MHC haplotypes. MAP VII also inducedthe production of significantly higher titres in the rodentsystem as seen in other immunization studies using rodentmalaria parasites (Tamet al. 1990, Migliorini et al. 1993,Wang et al. 1995). In addition, the lower anti-sporozoiteantibody titres displayed by all animals as compared to theanti-MAP titres may be explained in part by a lowersensitivity of the IFAT technique (Tamet al. 1990), butalso by the presence of anti-peptide antibodies directed topeptide conformations not present in the native CS protein.Although MAP systems are aimed at overcoming thegenetic restriction in humans,Aotus might not recognizehuman epitopes similarly or with the same affinity ashumans.

The fact that sporozoites react with antibodies induced byMAP containing p11 is very interesting. These MAPs possessthe AGDR sequence two tandem repeats (GDRADGQPA/GDRADGQPA/GDRADGQPA). This tetrapeptide is recog-nized bya monoclonal antibodycapable ofprotectingprimatesagainst parasite challenge (Charoenvitet al. 1991). Thisminimal epitope is also recognized by sera of theAotusimmunized with such MAPs and by sera of individualsfrom the same endemic area of Colombia (manuscript inpreparation, Are`valo-Herrera et al. 1996). As expected,neither the monoepitopic B nor T-cell based MAP induceda good antibody response indicating the need of T-B cell co-operation for an early and significant production of anti-bodies. Although p11 contains both B and T cell epitopes,when it was presented in the monoepitopic MAP I and II was

a poor immunogen indicating that presented in such structureit can not accomplish both functions efficiently. This might bedue to overlapping of both epitopes within the small sequenceincluded in the MAP. This dual function may however bepossible in natural conditions where the CS protein hasnumerous repeats that would allow the achievement of bothfunctions by separated regions of the repeat domain. Thestrong immune response induced in mice immunized with aP. falciparumCS based MAP (T1B)4 composed of the T-cellepitope DPNANPNVDPNANPNV and the B-cell epitope(NANP)3 both from the central repeat domain, may be infavour of this hypothesis as both epitopes, although com-posed of partially homologous sequences have the B and Tsequences clearly separated (Munesingheet al. 1991). Theinclusion of two T cell epitopes derived from theP. vivaxCSprotein in MAP VIII was less efficient in inducing antibodyresponse. It may be due to the possible antigenic interplaywhen more than one T -cell epitope is present in the peptideconstructs.

It is very important that several MAPs particularly VI andVII induced a significant production ofg-IFN in vitro. g-IFNhas been related to malaria protection by preventing the intra-cellular development of parasite liver forms (Schofieldet al.1987) and the parasite multiplication in the circulation inanimals immunized withP. falciparumMSP-1 (Herreraet al.1992). This cytokine seems to play its protective role throughthe induction of nitric oxide production which is capable ofdestroying intra-hepatic malaria parasites. Althoughg-IFNby itself is not capable of directly killing malaria asexualblood stagesin vitro (Herreraet al. 1992), it is proposedthat in vivo, IL-12 stimulates non-B and non-T cells toproduceg-IFN which induce the production of nitric oxidecapable of killing both liver stages (Sedegahet al. 1994,Stevensonet al. 1995) and problably malaria blood stages.MAP VI and VII induced both specific antibodies andg-IFNproduction, a condition that may be required for functionalimmunity in nature.

Since the anti-malarial immune response primed by thesevaccines is aimed at being boosted by further exposure toparasite, MAP VI could be a suitable candidate of choice forits use in endemic areas. However, MAP VII containing theTT epitope may be important for the priming of a strongimmune response that may be of benefit for non-immunepeople visiting malaria endemic regions (Panina-Bordignonet al. 1989, Valmoriet al. 1992). In both cases, P11 mightfurther play its dual role of B and T cell epitope, boosting theimmune response by later exposure to the parasite inindividuals immunized with these two constructs. Thesetwo MAP are currently being analysed in protection experi-ments in primates. If such experiments show significantprotective efficacy against homologous challenge, one maytake advantage of the possibility to include different malaria

S.Herreraet al. Parasite Immunology

q 1997 Blackwell Science Ltd,Parasite Immunology, 19, 161–170168

B-cell epitopes in a single immunogenic MAP construct(Ahlborg et al. 1995). A vaccine combining B-cell epitopescorresponding to the two described variants of theP. vivaxCS and/or other epitopes could be designed using the MAPsystems.

ACKNOWLEDGEMENTS

This investigation received financial support from theUNDP/WORLD BANK/WHO Special Programme forResearch and Training in Tropical Diseases (WHO/TDR).from the Fondo Colombiano de Investigaciones Cientı´ficasy Proyectos Especiales ‘Francisco Jose´ de Caldas’ (COL-CIENCIAS) and from the Fundacio´n para la Educacio´nSuperior (Fundacio´n FES). We would like to acknowledgethe technical assistance of Y.Zonenshine, G.Renjifo,C.Zapata and M.Salas. To Professor W.J.Pichier at theInstitut fur Immunologie und Allergologie. Inselspital,Bern, Switzerland for providing the primers used in theHLAtyping.WealsothankS.Belli for reading themanuscript.

Abbreviations: CS, Circumsporozoite; MAP, MultipleAntigen Peptides; PBMC, Peripheral Blood MononuclearCells; SI, Stimulation Index; CTL, Cytotoxic T Lympho-cytes; IFAT, Immuno-fluorescence antibody test; ELISA,Enzyme-Linked Immunosorbent Assay. TT, TetanusToxoid; tt, Tetanus toxin.

REFERENCES

Arnot D., Barnwell J.W. & Steward M.J. (1988) Does biased geneconversion influence polymorphism in the circumsporozoite protein-encoding gene ofPlasmodium vivax? Proceedings of the NationalAcademy of Sciences, USA85, 8102–8106

Atherton E., Loga C.J. & Sheppard R.C. (1988) Peptide synthesis. II.Procedures for solid phase synthesis using Na-fluorenymethoxy-carbamilamino-acids on polyamide synthesis of substance P and ofacyl-carrier protein 65-74 decapeptide.Journal of the ChemicalSociety Perkins Transaction1, 538–548

Ballou W.R., Hoffman S.L., Sherwood S.A.et al. (1987) Safety andefficacy of a recombinant DNAPlasmodium falciparumsporozoitevaccine.Lanceti, 1277–1281

Charoenvit Y., Collins W.E., Jones T.et al. (1991) Inability of malariavaccine to induce antibodies to a protective epitope within itssequence.Science251,668–671

Clyde D.F., McCarthy V.C., Miller R.M.et al. (1975) Immunization ofman againstPlasmodium falciparumandvivax by use of attenuatedsporozoites.American Journal of Tropical Medicine and Hygiene24,397–401

Currier, J., Sattabongkot J. & Good, M. (1992) ‘Natural’ T cells formalaria: evidence of highly sensitive malaria-specific responses fromnon-exposed donors.International Immunology4, 985–994

Del Giudice G., Grillot D., Re´nia L. et al.(1990) Peptide-primed CD4+cells and malaria sporozoites.Immunology Letters25, 59–63

Demotz S., Matricardi P., Lanzavecchia A.et al. (1989) A novel and

simple procedure for determining T cell epitopes in protein antigens.Journal of Immunological Methods122,67–72

Egan J.E., Weber J.L., Ballou W.R.et al. (1987) Efficacy of murinemalaria sporozoite vaccines: Implication for human vaccinedevelopment.Science236,435–456

Franke E., Lucas C.M., Chauca G.et al. (1992) Antibody response tothe circumsporozoite protein ofPlasmodium vivaxin naturallyinfected humans.American Tropical Medicine & Hygiene43,320–326

Gao, X., M. Fernandez-Vina, Shumway W.et al. (1990) DNA typingfor class II HLA antigens with allele-specific or group-specificamplification. Typing for subsets of HLA-DR4.Human ofImmunology27, 40–45

Gallati, H., I. Pracht., J. Schmidt.et al. (1987) A simple, rapid andlarge capacity ELISA for biologically active and recombinant humanIFN-g. Journal of Biology Regulation and Homeostatic Agents1,109–118

Herrera M.A., Rosero F., Herrera S.et al. (1992) Protection againstmalaria in Aotus monkeys immunized with a recombinant blood-stages antigen fused to an universal T-cell epitope: Correlation ofserum-Interferon levels with protection.Infection and Immunity60,154–158

Herrera S., Escobar P., De Plata C.et al. (1992) Human recognition ofT-cell epitopes on thePlasmodium vivaxcircumsporozoite protein.Journal of Immunology48, 3986–3990

Herrera M.A., De Plata C., Gonzalez J.M.et al.(1994) Immunogenicityof Multiple Antigen Peptides (MAPs) containingPlasmodium vivaxepitopes in BALB/c mice.Memorias do Instituto Oswaldo Cruz89,69–73

Herrington D.A., Clyde D.F., Losonsky G.et al. (1987) Safety andimmunogenicity in man of a synthetic peptide malaria vaccineagainstPlasmodium falciparumsporozoite.Nature328,257–25

Hollingdale M.R., Nardin E.H., Tharavany S.et al. (1984) Inhibitionof entry of Plasmodium falciparumand P. vivax sporozoites intocultured cells, anin vitro assay of protective antibodies.Journal ofImmunology132,909–913

Khusmith S., Charoenvit Y., Kumar S., Sedegah, M.et al. (1991)Protection against malaria by vaccination with sporozoite surfaceprotein 2 plus CS protein.Science252,715–718

Kumar S., Miller L.H., Quakyi I.A.et al. (1988) Cytotoxic T cellspecific for the circumsporozoite protein ofPlasmodium falciparum.Nature334,258–260

Lopez J.A., Gonza´lez J.M., Kettner A.et al. (1997) Synthetic peptidescorresponding to the non-repeat regions from the circumsporozoiteprotein of Plasmodium falciparum: human T-cell recognition andimmunogenicity in owl monkeys.Annals of Tropical Medicine andParasitology(in press)

Mann V.H., Huang, T., Cheng Q.et al. (1994) Sequence variation in thecircumsporozoite protein gene ofPlasmodium vivaxappears to beregionally biased.Molecular Biochemical Parasitology68, 45–52

McCarthy V.C. & Clayde D.F. (1977)Plasmodium vivax: correlation ofcircumsporozoite precipitation (CSP) reaction with sporozoite-induced protective immunity in man.Experimental Parasitology41, 167–171

Migliorini P., Betschart B. & Corradin, G. (1993) Malaria vaccine:immunization of mice with a synthetic T cell helper epitope aloneleads to protective immunity.European Journal of Immunology23,582–585

Munesinghe D.Y., Clavijo P., Calvo-Calle J.M.et al. (1991) Antigeni-city of multiple antigen peptides (MAP) containing B and T cellepitopes from the central repeat region of theP. falciparumcircum-sporozoite proteinEuropean Journal of Immunology21,3015–3020

Volume 19, Number 4, April 1997 CSP. vivaxbased Multiple Antigen Peptides

q 1997 Blackwell Science Ltd,Parasite Immunology, 19, 161–170 169

Nussenzweig R., Vanderberg J. & Most, H. (1967) Protective immunityproduced by the injection of x-irradiated sporozoite ofPlasmodiumberghei. Nature216,160–162

Panina-Bordignon P., Tam A., Termijtelen A.et al. (1989) Universallyimmunogenic T cell epitopes; promiscuous binding to human class IIand promiscuous recognition by T cells. European Journal ofImmunology19, 2237–2242

Potocnjak P., Yoshida N., Nussenzweig R.S.et al. (1980) Monovalentfragments (Fab) of monoclonal antibodies to a sporozoite surfaceantigen (Pb44) protect mice against malaria infection.Journal ofExperimental Medicine151,1504–1513

Romero P., Maryanski J.L., Corradin G.et al. (1989) Cloned cytotoxicT cells recognize an epitope in the circumsporozoite protein andprotect against malaria.Nature341,323

Rosenberg R., Wirtz R.A., Lanar D.et al. (1989) Circumsporozoiteprotein heterogeneity in the human malaria parasitePlasmodiumvivax. Science245,973–976

Schofield L., Ferreira A., Altszuler R.et al. (1987) Interferon-gammainhibits the intrahepatocytic development of malaria parasitein vitro.Journal of Immunology139,2020–2023

Sedegah M., Finkelman F. & Hoffman S. (1994) Interleukin 12 inductionof interferong-dependent protection against malaria.Proceedings ofthe National Academy of Sciences, USA.91, 10700–10702

Sinigaglia F., Guttinger M., Kilgus J.et al. (1988) A malaria T-cellepitope recognized in association with most mouse and human MHCclass II molecules.Nature336,778–780

Stevenson M., Tam M.F., Stanley F.W.et al (1995) IL-12-inducedprotection against blood-stagePlasmodium chabaudiAS requiresIFN-g and TNF-a and occurs via a nitric oxide-dependentmechanism.Journal of Immunology155,2545–2556.

Tam J.P. (1988) Synthetic peptide vaccine design: Synthesis andproperties of a high-density multiple antigen system.Proceedingsof the National Academy of Sciences, USA85, 5409–5413

Tam J.P., Clavijo P., Lu Y.A.et al. (1990) Incorporation of T and Bepitopes of the circumsporozoite protein in a chemically definedsynthetic vaccine against malaria.Journal Experimental Medicine171,299–306

Valmori D., Pessi A., Bianchi E.et al. (1992) Use of human universallyantigenic tetanus toxin T-cell epitopes as carriers for humanvaccination.Journal of Immunology149,717–721

Wang R., Charoenvit Y., Corradin G.et al. (1995) Induction ofprotective polyclonal antibodies by immunization with aPlasmo-dium yoelii circumsporozoite protein multiple antigen peptidevaccine.Journal of Immunology154,2784–2793

Weiss W.A., Mellouk S., Houghten R.A.et al.(1990) Cytotoxic T cellsrecognize a peptide from the circumsporozoite protein onmalaria infected hepatocytes.Journal Experimental Medicine171,763–773

Weiss W.R., Sedegah M., Beaudoin R.L.et al. (1988) CD8þ T cell(cytotoxic/supressors) are required for protection in mice immunizedwith malaria sporozoites.Proceedings of the National Academy ofSciences, USA85, 573–576

S.Herreraet al. Parasite Immunology

q 1997 Blackwell Science Ltd,Parasite Immunology, 19, 161–170170