Immunology Letters, 30 (1991) 75-80 Elsevier IMLET 01657

Multiple antigen peptides for specific detection of antibodies to a malaria antigen in human sera Annette HabluetzeP ,2, Antonello Pessi 3, Elisabetta Bianchi a, Gianfranco Rotigliano 4,5 and Fulvio Esposito 1 IDepartment of Molecular, Cellular and Animal Biology, Universit~ degli Studi di Camerino, Camerino (MC); 2World Health Organization Collaborating Centre for Malaria Epidemiology, Institute for Parasitology, Universitit "La Sapienza'~ Rome; 3Institute for Molecular Biological Research, Pomezia (Rome); 4Directorate Generalfor Development Aid, Ministry of Foreign Affairs, Rome; and 5Clinic for Tropical Diseases 3a Cattedra, Universitit "La Sapienza'; Rome, Italy (Received 21 May 1991; accepted 31 May 1991)

1. Summary

Multiple antigen peptides (MAP), consisting of a number of peptide copies synthesized on a branching lysyl core, offer a novel approach for rendering small peptides immunoreactive in solid-phase immunoassays. An octameric MAP, carrying 6 repeats of the sequence -N-A-A-G-, tandem repeated in the immunodominant region of the circumsporozoite (CS) protein of Plasmodium malariae, was used as a model to evaluate the suitability of the MAP system in an indirect enzyme-linked immunosorbent assay (ELISA) for detecting antibodies against a parasite antigen in individuals exposed to natural infection. The reaction of endemic sera in ELISA on MAPs-(NAAG)6 was related to that obtained in immunofluorescence on sporozoites, indicating the specificity of the antibody-MAP interaction. The reactivity of immune sera was found to be directed only against the (NAAG)6 moiety of the MAP and not against the lysyl core, since antibody binding to MAPs-(NAAG)6 was completely inhibited by (NAAG)6-NA monomer, but remained uninfluenced when lysyl core was used as competing ligand. The levels of antibodies to MAPsKey words: Multiple antigen peptide; ELISA; Malaria, Plasmodium malariae; Immunoassay Correspondence to: Annette Habluetzel, Dipartimento di Biologia Molecolare, Cellulare e Animale, 62032 Camerino (MC), Italy. Tel.: (737)-40738; Fax (737)-40764.

(NAAG)6 , in two groups of individuals naturally exposed to malaria infection, appeared to be related to their respective exposure to the parasite. 2. Introduction

Sporozoites, the infective stage of malaria parasites, are surrounded by a surface coat which is constituted mainly by a single type of molecule, the circumsporozoite (CS) protein. The central part of this protein is characterized by tandem repeated sequences, which have been demonstrated to represent an immunodominant B-epitope [1]. Antibodies to this epitope have been measured in individuals exposed to natural infection, employing immunoassays based on different antigen formulations: recombinant products of the CS gene [2, 3] and short synthetic peptides, either coupled to a protein carrier [4, 5] or polymerized [6, 7], since small peptides are weak immunoreagents in heterogeneous assays. Recently, a novel approach has been described to improve the immunoreactivity of small peptides: the multiple antigen peptides [8]. These chemically and structurally defined synthetic constructs consist of a branching lysyi core matrix on which multiple copies of peptide epitopes are synthesized. Tam and Zavala [9] demonstrated the specific reactivity of multiple antigen peptides (MAP) in immunoassays, with model molecules carrying as antigen the repetitive epitope of the CS protein of Plasmodium falciparura or Plasmodium berghei. These MAP react-

0165-2478 / 91 / $3.50 © 1991 Elsevier Science Publishers B.V. All rights reserved.

75

ed with monoclonal antibodies raised against sporozoites of the relevant species and with polyclonal antisera from mice immunized with the respective synthetic peptides. This study was designed to clarify whether the MAP system would be suitable for detecting antibodies in naturally immunized individuals, i.e., whether the antibody reactivity against the native antigen would correspond to that against a MAPconstruct containing the same antigenic sequence. This issue is dealt with here, using as a model an octameric MAP carrying 6 repeats of the sequence -NA-A-G-, which is tandem repeated in the immunodominant region of the CS protein of P malariae. 3. Materials and Methods

3.1. Synthesis of MAP8-(NAAG) 6 The synthesis of MAPa-(NAAG)6 has been described in detail elsewhere [10]. Briefly, the sequence was assembled by the flow-polyamide method [11], on a resin functionalized at a level of 0.02 mEq/g, which is lower than the usual one. A single coupling protocol was used, with a 2.5 M excess of Fmoc-amino acids (pentafluorophenyl esters) in the presence of N-hydroxybenzotriazole. The octameric lysyl core was assembled by sequential couplings of Fmoc-Lys (Fmoc), with standard cycles. To obtain the core for inhibition experiments, an aliquot of the resin was removed at this stage and acetylated in batch. The synthesis was performed on an automated Biolynx 4170 synthesizer (Pharmacia-LKB) and the assembly was monitored by the usual color tests. The assembly was straightforward: standard acylation kinetics was observed throughout, without any need for recoupling. This indicated that this method is a suitable alternative to the Boc-Benzyl approach, described by Tam [8] for MAP synthesis. The quality of the final product was checked by amino acid analysis and high pressure liquid chromatography in 2 different systems - reversed phase and gel permeation - both showing a single, quasi~gaussian peak. 3.2. Monoclonal antibodies and human sera Two monoclonal antibodies (mAb) were available: mAb-M, raised against P.. malariae sporozoites 76

(kindly provided by Frank Collins, CDC Atlanta) and mAb-B, raised against P brasilianum sporozoites (kindly provided by Alan Cochrane and Fidel Zavala, New York University). The CS proteins of P. brasilianum and P malariae share the tandem repeated sequence -N-A-A-G- [12, 13]. Because, in a competition experiment, mAb-M and mAb-B recognized the same epitope on MAPs-(NAAG)6 (data not shown), only mAb-B was used in the experiments reported below. Endemic sera originated from 2 groups of 6-36-month-old children, living in an area of Burkina Faso (West Africa) which is hyperendemic for falciparum malaria, and in which mixed infections with P.. malariae occurred with a frequency of up to 20% (E Esposito, S. Lombardi and D. Modiano, unpublished observations). Non-endemic sera (N=34) were obtained from Italian donors. 3.3 MAP8-(NAAG)6-based indirect enzyme-linked immunosorbent assay (ELISA) For detecting antibodies to MAP8-(NAAG)6, a standard ELISA protocol was used [14], with minor modifications. Medium affinity microtiter plates (Polysorp, Nunc, Denmark) were coated overnight with 100/zl/well of MAPs-(NAAG)6 at a concentration of 1/zg/ml in 0.15 M phosphate-buffered saline (PBS) pH 7.2, containing 0.05070 (w/v) sodium azide (PBS-Az). After washing with Tween 20, 0.05 070(v/v) in PBS (WB), the plates were postcoated overnight with 140/~I/well of bovine serum albumin (BSA), 0.002070 (w/v) in PBS-Az. The plates were washed once with WB, three times with distilled water and desiccated under vacuum. Sera or mAb, diluted in PBS-Az containing 1070 (v/v) Tween 20 and 0.1°70 (w/v) BSA, were incubated in the wells for 1 h (in this and in all the subsequent steps, volumes of 100 tzl/well were added). After washing, the binding of serum antibodies was detected with alkaline phosphatase-conjugated goat antibodies to human immunoglobulins (cat. AP120A, Chemicon, Temecula, CA). The binding of mAb was detected with peroxidase-conjugated goat antibodies to mouse immunoglobulins (cat. 725301-1, Pel-Freez Biologicals, Rogers, AR). Both conjugates, diluted in PBS containing 1070Tween 20 and 0.1070 BSA, were incubated for 1 h at the concentrationindicatedbythe producer. After washing, the reaction was visualized with the respective enzyme substrate- chromogens: 1 mg/ml

p-nitrophenylphosphate (cat. N 9389 Sigma, St. Louis, MO) in 1.1 M diethanolamine buffer, pH 9.8, containing 0.6mM MgCI2, for phosphataseconjugated sera; 0.4 mg/ml o-phenylenediamine (cat. 523120, Calbiochem, La Jolla, CA) in 0.1 M citrate buffer, pH 4.75, containing 0.1% (w/v) urea peroxide (cat. 666121, Calbiochem, La Jolla, CA), for antibodies conjugated to peroxidase. The phosphatase reaction was stopped after 1 h with 25 #l/well of 3 M NaOH, and absorbance values were read at 405 nm; the peroxidase reaction was stopped after 20 min with 25 #l/weU of 2 M H 2 8 0 4 , and absorbance values were read at 492 nm. The whole procedure was conducted at room temperature (25 + 5 oc).

wells at serial dilutions and the mAb then added at a fixed concentration. As shown in Fig. 1, the synthetic construct reacted efficiently with the mAb, still giving an A492 of about 1.0 at a concentration of 30 ng/ml. With the (NAAG)6-NA monomer, a 65 times higher peptide concentration was needed to obtain the same A492. This result confirmed the expected poor immunoreactivity of the small monomer in the assay and demonstrated that the epitope of the CS repeat region was still well presented by the MAP construct, after its binding to the microtiter plate. 4.2. Comparison of serum reactivity in ELISA on MAPs-(NAAG) 6 and in immunofluorescence on P. brasilianum sporozoites

4. Results and Discussion

4.1. Reactivity of plate-bound M A P r ( N A A G ) 6 with mAb-B The antigenicity of plate-bound MAPs-(NAAG) 6 was compared with that of the (NAAG)6-NA monomer, using mAb-B, raised against P. brasilianum sporozoites. The two peptides were coated on the microtiter optical density

Sera originating from individuals living in a P. malariae endemic area of Burkina Faso were preliminarily assayed with the MAPa-(NAAG)6ELISA. Ten of these sera, the absorbance values of which ranged from 0.05 to > 2.00, were tested in indirect immunofluorescence (IFAT) on air-dried P brasilianum sporozoites. The intensity of the TABLE 1 Comparison of serum reactivity in MAPs-(NAAG)6-based ELISA and in IFAT on air-dried P. brasilianum sporozoitesa

2-

Sera 1.5

ELISA b

IFAT c

$3 s S

s

0.5

0

I

I

I

I

I

I

0.002

0.008

0.032

0.125

0.5

2

1:50

I"

peptide concentration (~g/ml) Fig. 1. Reactivity of plate-bound MAPs-(NAAG)6 (solid squares, solid line) and (NAAG)6-NA monomer (open squares, dashed line), tested with a monoclonal antibody (mAb-B) directed against P.. brasilianum sporozoites. The two peptides were 4fold serially diluted - starting at a concentration of 8 #g/ml and coated on the microtiter wells. The mAb-B was added at the concentration of 4 #g/ml and the reaction was detected with peroxidase-conjugated antibodies to mouse immunoglobulins. The plotte d optical densities represent the median of triplicate wells.

K514 M20 K147 Z44 K504 K98 M70 M182 M335 Z340 cont. 1 cont. 2

2.34 2.22 0.82 0.73 0.72 0.33 0.24 0.13 0.10 0.05 0.05 0.10

+ + + + + +

1:100 + + + + + + + -

+ + + + + + -

a Sporozoite-coated slides were a generous gift of Fidel Zavala. b The values represent the median optical density of triplicate wells, c Endemic and non-immune control sera were both diluted 1:50 and l:100 in PBS containing 0.5% BSA, and incubated on air-dried, P. brasilianum sporozoite-coated slides. The reaction was detected with fluoresc¢in-labelled goat anti-human immunoglobulins (cat. API20F, Chemicon, Temecula, CA).

77

fluorescent reaction observed with each serum, was in agreement with the respective A492 value obtained in the MAPs-(NAAG)6-ELISA (Table 1). This result indicated that the synthetic construct was recognized by the antibodies reacting with P.. malariae sporozoites. 4.3. Competition mAb-B

between

endemic

sera

and

A competition experiment with endemic sera and mAb-B was set up, to verify the specificity of the serum antibody reaction with the M A P construct. A pool of endemic sera, strongly reactive on M A P 8(NAAG)6, was incubated on the antigen coated plate in serial dilutions. The mAb-B was then added at a fixed concentration and its binding detected by an anti-mouse immunoglobulin conjugate. As illustrated in Fig. 2, the reactivity of mAb-B was inversely correlated with the concentration of the serum pool and a complete inhibition of mAb-B binding was obtained at a pool dilution of 1:2. This demonstrated that the monoclonal antibody and the antibodies contained in the endemic sera recognized the same epitope on MAP8-(NAAG) 6.

4.4. (NAAG)6-NA moiety-restricted binding to MAP8-(NAAG) 6

antibody

A further competition experiment was designed, in order to assess whether the binding of antibodies contained in endemic sera was confined to the (NAAG)6-NA moiety of the synthetic construct or whether the lysyl core was also involved in the reaction. From a group of MAP8-(NAAG) 6 reactive sera, five were selected which gave A492values > 0.3 on a plate coated with the lysyl core moiety only. These potentially core binding sera were assayed on MAPs-(NAAG) 6, in the presence of lysyl core at a concentration of 200 # g / m l or of serial dilutions of (NAAG)6-NA monomer. The lysyl core did not affect serum reactivity, whereas (NAAG)6-NA monomer, at the same concentration, completely inhibited the antibody binding to the MAP-construct (Fig. 3). A considerable decrease of antibody binding was already achieved with a concentration of (NAAG)6-NA m o n o m e r of 3.2 #g/ml. In presence of this amount o f competing ligand, the optical densities of the sera were reduced to 2 4 - 3 8 % of those obtained in the absence of competition (Fig. 3). optical density

optical density 1.5

1.5 1

0.5

0.5 0

I 0

I 0.05

i 02

I 0.8

I 3.2

I 12.5

I 50

I 200

~200

(NAAG)sNA 0Jg/ml) 4

8

16

32

64

128

256

reciprocal serum dilution Fig. 2. Inhibition of mAb-B binding to M A P s - ( N A A G ) 6 by endemic sera. A pool o f endemic sera, strongly reactive on the MAPconstruct, was 2-fold serially diluted - starting with 1:2 - and incubated in the antigen coated plate. The mAb-B was added at a concentration of 1 ttg/ml and the reaction visualized with a peroxidase-labelled anti-mouse immunoglobulin conjugate. The solid circle in the upper right corner of the graph indicates the optical density (OD) value obtained with mAb-B in absence of competing endemic sera. The plotted ODs represent the median of triplicate wells.

78

Fig. 3. Antibody binding to M A P s - ( N A A G ) 6 in presence of (NAAG)6-NA m o n o m e r or lysy! core. Five endemic sera (represented by different symbols), diluted 1:200, were incubated overnight with 4-fold serial dilutions of (NAAG)6-NA m o n o m e r or with 200 # g / m l of lysyl core, and then processed on the MAP8-(NAAG) 6 coated plate. Symbols on the right indicate the reactivity of the corresponding serum in the presence of 200 # g / m l of lysyl core. The dashed line is placed at the height of the optical density (OD) value obtained with a pool of noni m m u n e sera. The plotted ODs represent the median of triplicate wells. Similar results have been obtained in mice immunized with M A P s - ( N A A G ) 6 by Del Giudice et al. [15].

These results indicated that the reactivity of the sera was exclusively directed to the antigen moiety of MAPs-(NAAG) 6. 4.5. Reactivity of sera from children of two endemic villages Antibodies of MAPs-(NAAG)6 were measured in children living in two villages of Burkina Faso, characterized by different intensity of P. malariae transmission (plasmodic indices for P. malariae: 11070 and 31o7o in village 1 and 2, respectively). As shown in Fig. 4, the levels of antibodies observed in the two villages, reflected the respective plasmodic index for P. malariae. To what extent these antibody levels are correlated directly with the frequency of sporozoite inoculations or reflect cross-reactivity against P.. malariae blood stage antigens [16], remains to be clarified. However this result provided evidence that antibodies naturally raised against P. malariae were specifically detected in the MAPbased ELISA. 4.6. Concluding remarks In the field of malaria vaccine research, much effort is concentrated on the characterization of parasite proteins which may elicit protective i m m u n e mechanisms in humans living in endemic areas. Immuno-epidemiological studies with synthetic

Acknowledgements

optical density 1.96 1.44 1.00 0.64 0.36 0.16

•

=i

peptides are being conducted to identify T-helper, Tcytotoxic and B-cell epitopes, with the aim of defining the components of a subunit vaccine. For measuring antibody responses to peptides representing parasite epitopes, immunoassays are being used, usually conjugating the peptide to a protein carrier or polymerizing it, to increase its immunoreactivity. The MAP-system may be advantageous compared with these conventional methods, as it provides a chemically and structurally unambiguous molecule, of which more than 80070 is represented by the antigen-moiety, thus rendering reactions with the core-moiety very unlikely. These theoretical considerations were experimentally confirmed in this study, using an octameric MAP, which carries 6 repeats o f the sequence -N-AA-G-, tandem repeated in the CS protein of P. malariae. In particular, the reactivity o f endemic sera with MAPs-(NAAG) 6 corresponded to that of the same sera in IFAT on sporozoites; the antibody binding to the MAP - completely inhibited by (NAAG)6-NA monomer - remained uninfluenced by the lysyl core and, finally, the levels of antibodies to MAP 8(NAAG)6 in children exposed to P. malariae infection were seen to be related to the circulation of the parasite in the area. Thus, the MAP system, applied in a solid-phase immunoassay, revealed to be suitable for detecting antibodies in sera from naturally immunized individuals.

•

•

•

•

•

•

•

Am

Am

~

•

0.04 village 1

village 2

Fig. 4. Reactivityof sera from children of two villages situated in a malaria endemic area. Sera were processed in the MAP8(NAAG)6-based ELISA at dilutions of 1:200. Village 1 and village 2 werecharacterizedby differentlevelsofR malariaeparasite prevalences: llO/0 and 31070,respectively.The ordinate of the dashed line correspond to the mean optical density (OD) of 35 sera from non-immune donors. The plotted ODs represent the median of triplicate wells.

We wish to acknowledge the helpful comments of F. Zavala on the experimental design, the excellent technical assistance of R. GambeUa and D. Modiano, and the generous collaboration o f the personnel of the "Centre de Lutte contre le Paludisme" in Ouagadougou. The activities described in this paper were supported by the Italian Ministry for Foreign Affairs - Directorate General for Cooperation in Development, the Italian C.N.R., the Italian Ministry for University and Scientific Research, the International Atomic Energy Agency, and the "Istituto Pasteur - Cenci Bolognetti" Foundation.

References [1] Zavala,E, Tam, J. P., Hollingdale,M. R., Cochrane,A. H., Quakyi, I., Nussenzweig,R. S. and Nussenzweig, V. (1985) 79

Science 228, 1436. [2] Hoffman, S. L., Wistar, R. Jr., Ballou, W. R., Hollingdale, M. R., Wirtz, R. A., Schneider, I., Marwoto, H. A. and Hockmeyer, W. T. (1986) New Engl. J. Med. 315, 601. [3] Hoffman, S. L., Oster, C. N., Plowe, C. V., Woollett, G. R., Beier, J. C., Chulay, J. D., Wirtz, R. A., Hollingdale, M. R. and Mugambi, M. (1987) Science 237, 639. [4] Esposito, E, Lombardi, S., Modiano, D., Zavala, F., Remme, H., Lamizana, L., Coluzzi, M. and Nussenzweig, R. S. (1986) Parassitologia 28, 101. [5] Esposito, E, Lombardi, S., Modiano, D., Zavala, E, Remme, H., Lamizana, L., Coluzzi, M. and Nussenzweig, R. S. (1988) Trans. R. Soc. Trop. Med. Hyg. 82, 827. [6] Tanner, M., Del Giudice, G., Betschart, B., Biro, S., Burnier, E., Degr6mont, A. A., Engers, H. D., Freyvogel, T. A., Lambert, P.-H., Pessi, A., Speiser, E, Verdini, A. S. and Weiss, N. (1986) Mem. Inst. Oswaldo Cruz 81, 199. [7] Del Guidice, G., Verdini, A. S., Pinori, M., Pessi, A., Verhave, G. P., Tougne, C., Ivanoff, W., Lambert, P.-H. and Engers, H. D. (1987) J. Clin. Microbiol. 25, 91.

80

[8] Tam, J. P. (1988) Proc. Natl. Acad. Sci. USA 85, 5409. [9] Tam, J. E and Zavala, E (1989) J. lmmunol. Methods 124, 53. [10] Pessi, A., Bianchi, E., Bonelli, F. and Chiappinelli, L. (1990) J. Chem. Soc. Chem. Commun. 8. [11] Dryland, A. and Sheppard, R. C. (1986) Peptide synthesis. Part 8. J. Chem. Soc. Perkin 1, 125. [12] Lal, A. A., de la Cruz, V. F., Collins, W. E., Campbell, G. H., Procell, P. M. and McCutchan, T. E (1988) J. Biol. Chem. 263, 4318. [ 13] Lal, A. A., de la Cruz, V. E, Campbell, G. H., Procell, P. M., Collins, W. E. and McCutchan, T. F. (1988) Mol. Biochem. Parasitol. 30, 291. [14] Engvall, E. and Perlmann, P. (1972) J. Immunol. 109, 129. [15] Del Giudice, G., Tougne, C., Louis, J. A., Lambert, P.-H., Bianchi, E., Bonelli, E, Chiappinelli, L. and Pessi, A. (1990) Eur. J. Immunol. 20, 1619. [16] Cochrane, A. H., Uni, S., Maracic, M., di Giovanni, L., Aikawa, M. and Nussenzweig, R. S. (1990) Bull. WHO 68 (Suppl.), 181-183.