Comparison of analytical methods for the evaluation of antibody responses against epitopes of polymorphic protein antigens

Journal of Immunological Methods 276 (2003) 19 – 31 www.elsevier.com/locate/jim

Comparison of analytical methods for the evaluation of antibody responses against epitopes of polymorphic protein antigens A. Helg a, M.S. Mueller a, A. Joss a, F. Po¨ltl-Frank a, F. Stuart b, J.A. Robinson b, G. Pluschke a,* b

a Swiss Tropical Institute, Socinstrasse 57, CH 4002, Basel, Switzerland Institute of Organic Chemistry, Winterthurerstrasse 190, University of Zu¨rich, CH-8057, Zu¨rich, Switzerland

Received 8 July 2002; received in revised form 31 January 2003; accepted 4 February 2003

Abstract Surface exposed protein antigens of the malaria parasite Plasmodium falciparum frequently harbor multiple dimorphic amino acid positions. These are associated with parasite immune evasion and represent a major obstacle for subunit vaccine design. Here, we have analyzed the flexibility of the humoral immune response against a semiconserved sequence (YX44LFX47KEKMX52L) of the key malaria blood stage vaccine candidate merozoite surface protein-1 (MSP-1). Monoclonal antibodies (mAbs) raised against one of the six described natural sequence variants of MSP-143 – 53 were analyzed for crossreactivity with the other allelic forms, which differ in one to three positions from the immunizing sequence. Enzyme-linked immunosorbent assay (ELISA) and surface plasmon resonance (SPR) spectroscopy demonstrated marked differences in mAb binding avidity to the variant sequences and isothermal titration calorimetry (ITC) provided evidence for a very low affinity of some of the interactions. In immunofluorescence analysis (IFA) and Western blotting analysis, the mAbs nevertheless stained all analyzed parasite clones expressing MSP-143 – 53 variant sequences. When used for the evaluation of humoral immune responses in clinical malaria vaccine trials, these two commonly used methods may thus not be suitable to distinguish biologically functional high affinity antibody responses from irrelevant low-affinity cross-reactivities. D 2003 Elsevier Science B.V. All rights reserved. Keywords: Antibody binding; Malaria; Antigenic polymorphism; Affinity constant determination

1. Introduction Abbreviations: IFA, immunofluorescence analysis; ITC, isothermal titration calorimetry; rMSP-1, recombinant merozoite surface protein-1; SPR, surface plasmon resonance spectroscopy; ELISA, enzyme-linked immunosorbent assay; mAb, monoclonal antibody; PCR, polymerase chain reaction; Ig, immunoglobulin; SDS, sodiumdodecylsulfate. * Corresponding author. Tel.: +41-61-284-8235; fax: +41-61271-8654. E-mail address: [email protected] (G. Pluschke).

It is now generally assumed that a highly effective subunit vaccine against Plasmodium falciparum malaria has to consist of a combination of several key antigens (Engers and Godal, 1998; Richie and Saul, 2002) or epitopes (Patarroyo et al., 1987; Shi et al., 1999) derived from different stages of the life cycle of this protozoan parasite. One of the leading blood stage vaccine candidate antigens is the mero-

0022-1759/03/$ - see front matter D 2003 Elsevier Science B.V. All rights reserved. doi:10.1016/S0022-1759(03)00075-9

20

A. Helg et al. / Journal of Immunological Methods 276 (2003) 19–31

zoite surface protein-1 (MSP-1). MSP-1 is synthesized during schizogony as a 190 –200-kDa glycoprotein, and is subsequently proteolytically processed into a range of defined fragments (Holder and Freeman, 1984). Immunization with purified P. falciparum MSP-1 or portions of it has protected monkeys partially or completely from malaria after experimental challenge with homologous and/or heterologous parasites (Hall et al., 1984; Patarroyo et al., 1987; Siddiqui et al., 1987; Etlinger et al., 1991; Chang et al., 1996; Egan et al., 2000; Kumar et al., 2000). Protective immunity induced by immunization with MSP-1 derived polypeptides is thought to be primarily antibody dependent (Hui and Siddiqui, 1987). Several parts of MSP-1 may elicit protective immune responses. Monoclonal antibodies for either a variant epitope in the N-terminal Block 2 or the C-terminal 19 kDa fragment have been shown to inhibit parasite growth in vitro (Blackman et al., 1990; Locher et al., 1996). Partial sterilizing protection was observed in Saimiri monkeys immunized with a peptide comprising residues 24 – 67 of Blocks 1 and 2 of MSP-1 (Cheung et al., 1986), and immunization with a peptide comprising residues 43 – 53 delayed the onset of disease in Aotus monkeys (Patarroyo et al., 1987). Like most of the other major malaria vaccine candidate proteins, MSP-1 contains multiple dimorphic amino acid positions, and individuals in endemic areas often harbor multiple infection with antigenic highly diverse parasite clones. The same sequence dimorphisms are found in all polyclonal parasite populations worldwide. They are thought to be associated with parasite immune evasion and represent one major obstacle to malaria subunit vaccine development. This has raised the question, how conserved an epitope has to be to serve as a suitable vaccine component? In order to understand the precise role of antigen dimorphism in immune evasion, we have analyzed the flexibility of CD4 T cell immune responses against a semiconserved sequence stretch of the N-terminal block of MSP-1 and found that all human T cell lines and clones analyzed were sequence variant specific (Daubenberger et al., 2002). In this report, we have now investigated the flexibility of the humoral immune response against the same semiconserved MSP-143 – 53 sequence stretch. Six variants of this sequence, which contains three dimorphic positions

(YX44LFX47KEKMX52L) have been found in P. falciparum populations worldwide. These variants comprise the amino acid combinations S44-Q47-V52, S44-H47-I52, G44-H47-I52, G44-Q47-V52, G44-H47-L52, G44-H47-V52 (Miller et al., 1993; Jiang et al., 2000). The MSP-143 – 53 variant S44-Q47-V52 corresponds to the 83.1 building block of the synthetic peptide malaria vaccine SPf66 (Patarroyo et al., 1987, 1988). We have generated MSP-1 reactive mAbs from mice immunized with SPf66 and analyzed cross-reactivity of the antibodies with the different naturally occurring allelic forms of MSP-143 – 53 by Western blotting, immunofluorescence analysis (IFA), ELISA, isothermal titration calorimetry (ITC) and surface plasmon resonance (SPR) spectroscopy. Results obtained with the different techniques are compared, and implications for the choice of appropriate techniques for the assessment of humoral immune responses to semiconserved vaccine candidate antigens are discussed.

2. Materials and methods 2.1. Monoclonal antibodies and recombinant MSP-1 Generation of monoclonal anti-SPf66 antibodies from mice carrying human Cn light chain gene segments has been described elsewhere (Pluschke et al., 1998). C- and N-terminally truncated MSP-1 of P. falciparum clone MAD-20 recombinantly expressed in E. coli (rMSP-134 – 595) was a kind gift of B. Takacs. The SPf66 peptide (CGDELEAETQNVYAAPNANPYSLFQKEKMVLPNANPPANKKNAGC)n was a kind gift of M. Patarroyo. 2.2. IFA 12-Well multitest immunofluorescence microscopy slides (Flow Laboratories, Baar, Switzerland) were pretreated with 0.01% poly-L-lysine (Sigma, St. Louis, MO) for 30 min at room temperature and washed five times with RPMI basal salt medium (Gibco BRL). Erythrocytes from in vitro cultures (Matile and Pink, 1990) of P. falciparum clones MAD-20 (S44-Q47-V52), RO71 (G44-H47-I52), NF54 (G44-Q47-V52), IFA9.2 (G44-H47-V52) and K1 (S44-H47-I52) with about 10% parasitemia were washed two times in RPMI at room temperature. Brazil-608, the only P. falciparum clone

A. Helg et al. / Journal of Immunological Methods 276 (2003) 19–31

described so far to express the G44-H47-L52 allele (Miller et al., 1993), was not available for these analyses. The C-terminal sequence of MSP-1 of the other clones was reconfirmed by sequencing of polymerase chain reaction (PCR) products as described (Jiang et al., 2000). Cells were resuspended in RPMI and mixed with two volumes of a solution containing 4% para-formaldehyde and 0.1% Triton X-100. Droplets of 30 Al cell suspension were added to each well, incubated at room temperature for 30 min and washed five times with PBS. Wells were incubated for 15 min at room temperature with blocking solution containing 10% fatty acid-free BSA in PBS. Immunostaining was started by incubating the wells with 20 Al of serial dilutions of purified mAb in blocking solution for 1 h at room temperature in a humid chamber. After five washes with blocking solution, 20 Al of 5 Ag/ml Cy3conjugated affinity-pure F(abV)2 fragment goat antimouse IgG antibodies (Jackson Immuno Research Laboratories, West Grove, PA), diluted in blocking solution containing 0.01 mg/ml Hoechst No. 33258 (Sigma), were added to the wells and incubated for 1 h at room temperature. Afterwards, slides were washed five times with PBS, mounted with 90% (v/v) glycerol containing 0.1 M TrisCl, pH 8.0, and 2 mg/ml ophenylenediamine and covered with a cover slip. Antibody binding and DNA staining was assessed by fluorescence microscopy. 2.3. Western blot analysis Total parasite protein preparations were obtained as described (Matile and Pink, 1990) by saponin lysis of erythrocytes from parasite cultures of the strains also used for IFA. Briefly, cultured parasites were washed three times with serum-free RPMI medium. Pelleted infected red cells were lysed by mixing with a large volume (adjusted to 5% hematocrit) of 0.015% (w/v) saponin in 150 mM NaCl and 15 mM sodium citrate (pH 7.0) and incubated on ice for 20 min. Finally, the pelleted parasites were resuspended in three volumes of SSC buffer and stored at 80 jC before use. Parasite lysates were run on 10% gels and separated proteins were transferred electrophoretically to a nitrocellulose filter (Protean Nitrocellulose, BA 85, Schleicher and Schuell). Blots were blocked with PBS containing 5% milk powder and 0.1% Tween-20 overnight at 4 jC. The filter was cut into strips and then incubated for 2 h

21

at room temperature with serial dilutions of anti-MSP-1 mAb in blocking buffer. After washing the strips three times for 10 min with blocking buffer, they were incubated with a 1:3500 dilution of horseradish peroxidase conjugated goat anti-mouse IgG antibodies (BioRad, Reinach, Switzerland) for 1 h at room temperature. Finally, after washing, bands were visualized by addition of ECL substrate (Amersham Pharmacia Biotech, Freiburg, Germany) and exposition to Kodak Biomax light scientific imaging films according to the manufacturer’s recommendations. 2.4. ELISA ELISA microtiter plates (Immulon 4B; Dynatech, Embrach) were coated overnight at room temperature with 50 Al per well of a 10 Ag/ml solution of polymerized peptides (CGYSLFQKEKMVLGC)n, (CGYGLFQKEKMVLGC)n, (CGYGLFHKEKMILGC)n, or (CGYGLFHKEKMLLGC)n or SPf66 in PBS (0.15 M, pH 7.2). The plates were washed three times with deionized water containing 0.5% Tween-20 and then blocked with 5% milk powder in PBS for 1 h at 37 jC. After blocking, the plates were washed and then incubated with serial dilutions of mAb in PBS containing 0.05% Tween-20 and 0.5% milk powder for 1 h at 37 jC. The plates were washed and incubated with goat anti-mouse IgG (g-chain specific) conjugated to alkaline phosphatase (Sigma) at 0.3 Ag/ml in PBS containing 0.05% Tween-20 and 0.5% milk powder for 1 h at 37 jC. To each well, 50 Al phosphatase substrate solution (1 mg/ml p-Nitrophenylphosphate (Sigma) in a pH 9.6 bicarbonate buffer solution containing 0.02% MgCl2 (w/v)) were added, incubated at room temperature and the absorbency was read after appropriate time at 405 nm using a Titertek Multiscan MCC/340 reader (Labsystems, Helsinki, Finland). All assays were done twice in duplicates, typical results are shown. Coating of plates with monomeric peptides yielded similar, but less consistent, results. 2.5. Competitive ELISA ELISA microtiter plates (Immulon 4B; Dynatech, Embrach) were coated overnight at room temperature with 50 Al per well of a 5 Ag/ml solution of rMSP134 – 595containing the sequence variant S44-Q47-V52 in PBS (0.15 M, pH 7.2) with 1.8 mM urea. The

22

A. Helg et al. / Journal of Immunological Methods 276 (2003) 19–31

plates were washed three times with deionized H2O containing 0.5% Tween-20 and then blocked with 5% milk powder in PBS for 1 h at 37 jC. After blocking, the plates were washed as before. Subsequently mAbs (at a fixed half-saturating concentration) were incubated together with serial dilutions of monomeric variant peptides (YSLFQKEKMVL, YGLFQKEKMVL, YGLFHKEKMIL and YGLFHKEKMLL) in PBS containing 0.05% Tween-20 and 0.5% milk powder. Subsequently, mAb binding was analyzed using alkaline phosphatase-conjugated goat anti-mouse IgG (g-chain specific) antibodies as described above. 2.6. SPR spectroscopy SPR spectroscopy was performed using a BIAcore apparatus (Biacore, Uppsala, Sweden). The CM5 microsensor chip was activated by injection of 50 Al 0.05 M N-hydroxysuccinimide (Sigma) and 0.2 M Nethyl-NV-(3-diethylaminopropyl)-carbodiimide (Fluka, Buchs, Switzerland), followed by injection of 50 Al of a 100 Ag/ml solution of SPf66 in 10 mM Na-acetate buffer (pH 4.5) to obtain a high density surface. Unreacted ester groups on the chip were blocked afterwards by injection of 35 Al 1 M ethanolamine – HCl (pH 8.5). All steps were performed at a flow rate of 5 Al/min. To determine a mAb concentration, which was sensitive to competition, two-fold serial dilutions of mAb were tested at a flow rate of 10 Al/min. Subsequently, competition experiments were performed at a final mAb concentration of 25 nM (yielding Rmax values of up to 2500) with monomeric variant peptides (YGLFQKEKMVL, YGLFHKEKMIL and YGLFHKEKMLL) at concentrations of 5, 50 and 500 nM. All dilutions were done in flow buffer and after pre-incubation for several hours, mAb-peptide mixtures were injected at a flow rate of 10 Al/min for 3 min with a following dissociation phase of 5 min. Between sample injections, the surface was regenerated using 10 Al 0.1 M HCl containing 0.005% surfactant P20. 2.7. ITC Calorimetric titration experiments were performed using an MCS-ITC instrument (MicroCal, Northampton, MA). The sample cell (1.34 ml) was filled with a

mAb solution (typically 2 AM) in PBS. The injection syringe (nominal volume 250 Al) was filled with a peptide solution (typically 50 –100 AM) in PBS. The reference cell contained a solution of 0.01% sodium azide. Unpolymerised peptides (SPf66 monomer without terminal cysteines, YGLFQKEKMVL, YGLFHKEKMIL, YGLFHKEKMLL and YSLFHKEKMIL were used for the analyses. During the experiments, the sample solution was stirred by rotating at 400 rpm the injection syringe whose tip has the form of a paddle. After the baseline, stability was better than 0.1 Acal/s, 1 –2 Al of solution were injected to remove possible air bubbles at the syringe openings. The preinjection was followed by a succession of injections of constant volume, typically between 5 and 15 Al at constant time intervals. The time intervals between two consecutive injections, typically 250 s, allowed the heat signal to come back to baseline. The instrument was equilibrated with an external circulating bath at least 5 jC below the experimental temperature. Unless otherwise indicated, experiments were carried out at 25 jC. Prior to each experiment, sample cell and syringes were rinsed with water. After each experiment, sample cell and syringes were rinsed first with water, cleaned with 200 ml of 0.1% sodiumdodecylsulfate (SDS) solution (Merck, Darmstadt, Germany) and finally rinsed with at least 1 l of double distilled water. The isothermal titration curve was registered and analyzed using ORIGIN software (MicroCal) provided with the MCS-ITC instrument. 2.8. Sequence analysis of immunoglobulin j light chain rearrangements Total RNA was prepared from about 107 hybridoma cells with the RNeasy Total RNA Kit (Quiagen, Hilden, Germany). First strand cDNA synthesis with random primers and AMV reverse transcriptase and PCR amplification of IGKV-J-C sequences were carried out using the GeneAmp RNA PCR kit (Roche Molceular Systems, Branchbury, NY) according to the manufacturer’s recommendations. For PCR, combinations of primers specific for the human IGKC (CTCATCAGATGGCGGGAAGATGAAGACA) and mouse IGKV gene segments (GAAATTGT(G/T)CTCAC(C/ A)CA(G/A)TCTCC, GACATCCAGATGAC(C/ A)CAG(T/A)CT(C/A)C, GATATTGTGATGAC(C/

A. Helg et al. / Journal of Immunological Methods 276 (2003) 19–31

23

Fig. 1. Alignment of cDNA nucleotide sequences of functional n light chain rearrangements from the hybridoma cell lines 7.27, 7.2, 7.4, 7.13, 7.34 with the mouse germ-line IGKV14-111*01 sequence and from the hybridoma cell lines 7.19, 7.11 with the mouse germ-line IGKV1-110*01 sequence, respectively. Identities among sequences are shown as dots.

24

A. Helg et al. / Journal of Immunological Methods 276 (2003) 19–31

A ) C A G G ( C / A ) T, G AT G T T G T G AT G A C C C AAACTCC, A(A/G)(T/C)ATTGTGATGACCCAG(A/T)CTC) were used. PCR products of the expected size were purified using the QIAqick-spin PCR Product Purification Kit 250 (Quiagen), cloned and sequenced in both directions using an ABI PRISM 310 Genetic Analyzer (Perkin Elmer, Foster City, CA). For each hybridoma, DNA clones from at least two

independent PCR amplifications were analyzed to exclude artifacts. Sequences were analyzed using software coming from the server of IMGT (http://imgt. cines.fr:8104; initiator and coordinator: Marie-Paule Lefranc, Montpellier, France), the international ImMunoGeneTics database (Lefranc, 2001). The nomenclature used for gene segments follows Martinez and Lefranc (1998).

Fig. 2. Indirect immunofluorescence staining of mature P. falciparum schizonts by mAb 7.27 (right panel). Parasite DNA was stained using Hoechst 33258 dye (left panel). Binding to parasite clones expressing different MSP-1 Block 1 sequence variants was analyzed: (a) MAD-20 (S44-Q47-V52); (b) RO71 (G44-H47-I52); (c) NF54 (G44-Q47-V52); (d) IFA9.2 (G44-H47-V52) and (e) K1 (S44-H47-I52). No staining was observed with an isotype matched control mAb (not shown).

A. Helg et al. / Journal of Immunological Methods 276 (2003) 19–31

3. Results 3.1. Generation of mAb cross-reactive with P. falciparum MSP-143 – 53 B cell hybridoma clones that secrete anti-SPf66 IgG were generated from spleen cells of a mouse immunized with the synthetic peptide malaria vaccine SPf66. Eighteen of the 64 anti-SPf66 mAbs obtained exhibited reactivity with the SPf66 building block MSP-143 – 53 (YSLFQKEKMVL). Eleven of these 18 mAbs had more than 20 times higher ELISA titer against SPf66 than against the synthetic MSP-143 – 53 peptide sequence polymerized through C- and Nterminally added GC residues. The remaining seven

25

mAbs exhibited anti-MSP-143 – 53 titers that were only four to seven times lower than those against the SPf66 peptide. Cross-reactivity of the 18 MSP-143 – 53 peptide-binding mAbs with MSP-1 of P. falciparum clone MAD-20 (variant S44-Q47-V52) was tested by IFA and by Western blotting. Only those seven mAbs with high relative anti-MSP-143 – 53 titers stained blood stage parasites, yielding staining patterns characteristic for the cell surface antigen MSP-1. In Western blotting experiments with blood stage parasite saponin-lysates, these seven mAbs and two of the mAbs with low relative affinity to MSP-143 – 53 stained unprocessed MSP-1, and to a variable extent, also some of the proteolytic processing products of MSP-1 (data not shown).

Fig. 3. Western blot staining of unprocessed MSP-1 and some of its processing products by mAbs 7.19 and 7.27. Parasite blood stage lysates of P. falciparum clones MAD-20 (S44-Q47-V52) or RO71 (G44-H47-I52) were tested with serial two-fold dilutions (lanes 1 – 8) of purified mAbs starting with a concentration of 1000 ng/ml (lane 1).

26

A. Helg et al. / Journal of Immunological Methods 276 (2003) 19–31

The seven hybridomas 7.2, 7.4, 7.11, 7.13, 7.19, 7.27 and 7.34, which secreted mAbs that reacted in IFA with blood stage parasites, were selected for further analysis. All mAbs were of the IgG1 subclass and contained a chimeric human – mouse n light chain because they were derived from a mouse carrying a n constant region replacement mutation (Pluschke et al.,

1998). For the analysis of a potential clonal relatedness of these hybridomas, we cloned and sequenced the variable region encoding gene segments of their nchain transcripts. Two IGKV-IGKJ combinations were found: the five clones 7.2, 7.4, 7.13, 7.27 and 7.34 (V14 group) expressed an IGKV14-111-IGKJ2 transcript and the hybridomas 7.11 and 7.19 (V1 group) an

Fig. 4. (a) Binding pattern of mAbs 7.27 and 7.19 to sequence variants of MSP-143 – 53 in ELISA; letters refer to positions 44, 47 and 52, respectively. (b) Inhibition of ELISA binding of mAbs 7.27 and 7.19 to plates coated with rMSP-134 – 595 (sequence variant S-Q-V) by peptide sequence variants of MSP-143 – 53; letters refer to positions 44, 47 and 52, respectively. (c) Inhibition of binding of mAbs 7.27 and 7.19 at 25 nM to the sensor surface with immobilized SPf66 (sequence variant S44-Q47-V52) by peptide sequence variants of MSP-143 – 53 in BIAcore solution affinity assay; letters refer to positions 44, 47 and 52, respectively.

A. Helg et al. / Journal of Immunological Methods 276 (2003) 19–31

IGKV1-110-IGKJ5 combination. Identity of the V-J junctional regions (Fig. 1) demonstrated clonal relatedness of the spleen cells that are the progeny of the hybridomas belonging to the V14 or the V1 group, respectively. Differences in the nucleotide sequences were indicative for divergence of the progenitor B cells of the individual clonally related hybridomas by somatic hypermutation. Sequencing of products from at least two independent amplifications to exclude PCR artifacts reconfirmed all nucleotide differences found. Of the 25 base differences from the germline sequences observed in the case of the members of the V14 family, only seven were silent. Similarly, in the case of the two sequences of the V1 group, four of five base exchanges caused changes in the amino acid sequence of the immunoglobulin (Ig) light chain variable region. This is a strong indication for the selection of B cells with somatically mutated affinity maturated IgG. A maximum of eight differences in the deduced amino acid sequence was found in the case of members of the V14 group. Sequences of hybridomas 7.11 and 7.19 differed at four positions. In spite of these amino acid exchanges in the variable regions, no significant differences in the binding patterns of mAbs within one group to the MSP143 – 53 sequence variant polymers were detected in ELISA (data not shown). Therefore, one mAb from each group (mAb7.27 and mAb7.19) was selected for further analysis. 3.2. Cross-reactivity with natural sequence variants of MSP-143 – 53 IFA was done with erythrocytes infected with P. falciparum parasite clones expressing MSP-1 with the sequences S44-Q47-V52 (MAD-20), S44-H47-I52 (K1), G44-H47-I52 (RO71), G44-Q47-V52(NF54) and G44H47-V52 (IFA9.2). In contrast to isotype matched controls, both mAb7.27 (Fig. 2) and mAb7.19 (not shown) stained the blood stages of all parasite clones tested. Minimal antibody concentrations required were between 50 and 500 ng/ml for all mAb/parasite clone pairs. Both mAbs also reacted in Western blotting with MSP-1 of the same five parasite clones tested (data not shown). In titration experiments, minimal antibody concentrations required were slightly lower with MAb 7.19 ( V 50 ng/ml) than with mAb 7.27 ( z 50 ng/ml). mAb titrations with lysates of the parasite

27

clones with the two most diverse MSP-1 epitope sequences, RO71 (G44-H47-I52) and MAD20 (S44Q47-V52) are shown in Fig. 3. ELISA was done with plates coated with polymerized peptide sequence variants (CGYXLFXKEKMXLGC)n. mAb 7.27 exhibited strong binding to the sequence variant used for immunization (S44Q47-V52), low binding to G44-H47-L52 and no significant binding to G44-Q47-V52 and G44-H47-I52 (Fig. 4a). mAb 7.19 exhibited a similar binding pattern, except for a low reactivity with the variant G44-Q47V52 (Fig. 4a). Isotype matched control antibodies exhibited no significant binding to any of the peptides (not shown). In competition ELISA inhibition of mAb binding to recombinantly expressed MSP-134 – 595 containing the sequence variant used for immunization (S44-Q47V52) by unpolymerized MSP-143 – 53 peptides was analyzed (Fig. 4b). With both antibodies, the homologous S44-Q47-V52 variant peptide was the most efficient competitor. In the case of mAb7.27, the S to G exchange at position 44 significantly weakened the inhibitory activity. Binding to MSP-1 was, however, inhibited by all three heterologous peptide variants tested. In the case of mAb7.19, the variant peptide with an S to G exchange at position 44 (G44-Q47-V52) competed binding as efficiently as the homologous peptide, while two more divergent sequences tested (G44-H47-I52 and G44-H47-L52) had no significant inhibitory activity. Solution affinity SPR assays were performed using a sensor chip with a high-density immobilization surface of SPf66. The same unpolymerized competitor

Table 1 Reactivity of mAb 7.27 and mAb 7.19 with MSP-1 sequence variants MSP-1 variants S44-Q47-V52 G44-Q47-V52 G44-H47-I52 G44-H47-L52 S44-H47-I52

mAb 7.27

mAb 7.19

KA (10e7/M)

IFA

KA (10e7/M)

IFA

12.30 < 0.03 0.47 0.13 1.34

+ + + nd +

0.46 < 0.03 < 0.03 < 0.03 < 0.03

+ + + nd +

ITC binding affinity constants (KA) and staining of blood stage parasites in IFA are compared. nd: Not determined (no parasite strain carrying the MSP-1 sequence variant G44-H47-L52 was available).

28

A. Helg et al. / Journal of Immunological Methods 276 (2003) 19–31

peptides as for the competitive ELISA were used. Inhibition patterns with the variant sequences were in complete agreement with those of the competition ELISA; i.e. binding of mAb7.27 was inhibited by all three variant sequences (G44-Q47-V52, G44-H47-I52 and G44-H47-L52) tested, while only the variant with the single S to G exchange at position 44 inhibited binding of mAb7.19 (Fig. 4c).

Binding constants (KA) obtained by ITC measurements for the interaction of mAbs with unpolymerized peptides in solution are listed in Table 1. mAb7.27 exhibited a more than 20-fold higher affinity for the immunizing peptide sequence than mAb7.19 (12  107 M 1 versus 0.5  107 M 1). In the case of mAb7.27, one of the variant sequences (S44-H47-I52) bound with about 10-fold reduced affinity while the

Fig. 5. ITC profiles of the binding of mAb 7.27 to sequence variants of MSP-143 – 53; letters refer to positions 44, 47 and 52, respectively.

A. Helg et al. / Journal of Immunological Methods 276 (2003) 19–31

binding to the other variant sequences analyzed was either too low for signal detection (G44-Q47-V52) or for precise determination of KA (G44-H47-I52 and G44H47-L52) by ITC (Fig. 5). In the case of mAb7.19, only titrations with the immunizing sequence variant yielded a detectable signal (data not shown).

4. Discussion Humoral immune responses to protein subunit vaccine candidates are commonly analyzed by solid phase assays, such as ELISA, in which the proteins or peptides used as immunogen are adsorbed to a surface. In addition, the generation of antibodies that cross-react with the pathogen-expressed target protein is often assessed by IFA and Western blotting. Due to immune selection pressure, surface exposed antigens of pathogens are often polymorphic. Broad cross-reactivity of vaccine induced immune responses is therefore a critical feature of a comprehensive vaccine. Accordingly, methods are required that assess the cross-reactivity of antibody responses against the naturally occurring sequence variants. Here, we have analyzed the interaction of mAbs generated against one variant of the semiconserved sequence Y43SLFQKEKMVL53 of the key malaria vaccine candidate MSP-1 with naturally occurring sequence variants (S44-H47-I52, G44-H47-I52, G44Q47-V52, G44-H47-L52, G44-H47-V52) that are found in all polyclonal malaria populations worldwide. Use of well-characterized mAbs strongly facilitated the comparison of different analytical techniques in this article. In vaccine trials, the characterization of the humoral immune responses is further complicated by the diversity of fine specificities and affinities of the elicited polyclonal antibody populations. In IFA and Western blotting, mAbs did not discriminate between P. falciparum clones expressing different MSP-143 – 53 variants. In contrast, pronounced differences in the relative avidity of the mAbs to the different sequence variants were observed in ELISA with immobilized synthetic peptides. In solid phase assays binding interaction of antibodies with the solid phase bound antigen may be limited by steric or attractive interactions between the mAb molecules themselves (Nygren et

29

al., 1987; Nygren and Stenberg, 1989), blocking of binding sites by multivalent binding and rebinding (Neri et al., 1996) and the conformation of the antigens adsorbed to the surface (Underwood, 1993). These effects and variable dissociation rates (Stenberg and Nygren, 1989) complicate the determination of ‘functional affinities’ and therefore assays allow primarily qualitative comparisons of avidity (Mattes, 1997). In competition assays mass transport effects play no major role because the antibody –antigen interactions to be measured take place in solution. If the binding process is equilibrated during incubation time, the amount of mAbs free for binding to the immobilized antigen is determined by the equilibrium constant KA of the mAb to the competitor. Since measurement of concentrations of free mAb in solution is not affected by mass transport limitations (Nygren and Stenberg, 1989), competitive assays enable a quite precise qualitative comparison of binding avidities of mAbs (Neri et al., 1996). Here, peptide mediated inhibition of mAb binding to surface immobilized recombinant MSP-1 in inhibition ELISA or to the immunizing agent (the SPf66 synthetic peptide) in SPR spectroscopy yielded comparable results. The discriminatory power of both competition assays was higher than that of direct ELISA. Antibody binding to the target protein and a competing effect by homologous synthetic peptide antigens are regarded as key criterion for specificity (Savoca et al., 1991). To verify the obtained qualitative affinity rankings to the sequence variants, thermodynamic parameters of antibody – antigen interaction were determined quantitatively by ITC. Here, measurements are carried out in solution, thus avoiding the adsorption of reactants to a solid phase and eliminating limitations by mass transport or diffusion rates. ITC results correlated best with the avidity rankings obtained in the competitive assays. The discriminative power of ITC was higher than that of the competition assays and allowed for example to rank the interactions of mAb 7.27 with the variant G44-H47-I52 (KA = 0.47  107/M) and G44-H47-L52 (KA = 0.13  107/M). The discriminative power of competition assays was, however, still sufficient to discriminate between mAb 7.27 binding to variant G44-H47-I52 (KA = 0.47  107/M) and G44-Q47-V52 (KA < 0.03  107/M).

30

A. Helg et al. / Journal of Immunological Methods 276 (2003) 19–31

At variance with most other analytical approaches, ITC determines simultaneously enthalpy and entropy contributions of the binding interactions as well as the binding constant and stoichiometry (Pierce et al., 1999). Many other methods aim to measure the binding constant K only, which is directly associated to the free energy of binding (DG = RT ln K). As the free energy (DG = DH TDS) is composed of an enthalpy (DH) and an entropy (DS) term, microcalorimetry offers a completely different approach for the characterization of a binding process. The enthalpy DH is generally considered as an indicator of the changes in intermolecular bond energies, while the entropy DS is an indicator for the rearrangements of the solvent molecules during the binding process. As practically all binding interactions are accompanied by an uptake (endothermic) or a release (exothermic) of heat, titration microcalorimetry is universally applicable for the characterization of binding processes. Calorimetric analyses do not require marker molecules or intrinsic spectroscopic reporter groups, which can modify the analyzed interactions. In our analyses of the interactions between mAbs with synthetic peptide variants, the detection limits of KA for accurate measurements were estimated to be about 3  105/M (c = 1) as lower and 3  108/M (c = 1000) as upper limit (according to the equation c = KA  M  n (M = mean mAb concentration in cell: 1.8 AM; n = number of binding sites: 2; c = parameter ranging from 1 to 1000 for accurate measurements; Wiseman et al., 1989). Our results demonstrate that interactions with affinity constants below the lower limit for ITC determination were still strong enough to demonstrate binding in IFA and Western blotting. While ITC allows to quantitatively determine thermodynamic binding parameters, its routine use is limited by the large amounts of reactants required. Our results demonstrate that competitive assays, like inhibition ELISA and SPR spectroscopy, which require only small amounts of reactants, are suitable alternatives for the determination of relative binding avidities. In contrast, the discriminatory power of methods like IFA and Western blotting that are commonly used in the evaluation of immune responses in vaccine trials is limited. Functionally irrelevant lowaffinity antigen – antibody interactions may thus not be sufficiently differentiated from affinity matured immune responses.

Acknowledgements We are grateful to Dr. M. Mutz (Novartis, Basel, Switzerland) for enabling and supervising the ITC measurements. We thank Mr. Luyong Jiang for preparing the MSP-143 – 53 variant peptide G44-H47V52 and Denise Vogel for an excellent technical assistance. SPf66 and its building block peptides were kindly donated by Professor M. Patarroyo (Instituto de Inmunologia, Bogota, Columbia). The recombinantly expressed N-terminal part of MSP-1 was obtained from Dr. B. Takacs (F. Hoffmann La Roche, Basel, Switzerland).

References Blackman, M.J., Heidrich, H.G., Donachie, S., McBride, J.S., Holder, A.A., 1990. A single fragment of a malaria merozoite surface protein remains on the parasite during red cell invasion and is the target of invasion-inhibiting antibodies. J. Exp. Med. 172, 379. Chang, S.P., Case, S.E., Gosnell, W.L., Hashimoto, A., Kramer, K.J., Tam, L.Q., Hashiro, C.Q., Nikaido, C.M., Gibson, H.L., Lee-Ng, C.T., Barr, P.J., Yokota, B.T., Hut, G.S., 1996. Recombinant baculovirus 42-kilodalton C-terminal fragment of Plasmodium falciparum merozoite surface protein 1 protects Aotus monkeys against malaria. Infect. Immun. 64, 253. Cheung, A., Leban, J., Shaw, A.R., Merkli, B., Stocker, J., Chizzolini, C., Sander, C., Perrin, L.H., 1986. Immunization with synthetic peptides of a Plasmodium falciparum surface antigen induces antimerozoite antibodies. Proc. Natl. Acad. Sci. U. S. A. 83, 8328. Daubenberger, C.A, Nickel, B., Ciatto, C., Gru¨tter, M.G., Po¨ltlFrank, F., Rossi, L., Siegler, U., Robinson, J., Kashala, O., Patarroyo, M.E., Pluschke, G., 2002. Amino acid dimorphism and parasite immune evasion: cellular immune responses to a promiscuous epitope of Plasmodium falciparum merozoite surface protein-1 displaying dimorphic amino acid polymorphism are highly constrained. Eur. J. Immunol. 32, 3667. Egan, A.F., Blackman, M.J., Kaslow, D.C., 2000. Vaccine efficacy of recombinant Plasmodium falciparum merozoite surface protein 1 in malaria-naive, -exposed, and/or -rechallenged Aotus vociferans monkeys. Infect. Immun. 68, 1418. Engers, H.D., Godal, T., 1998. Malaria vaccine development: current status. Parasitol. Today 14, 56. Etlinger, H., Caspers, P., Matile, H., Scho¨nfeld, H., Stu¨ber, D., Tacacs, B., 1991. Ability of recombinant or native proteins to protect monkeys against heterologous challenge with Plasmodium falciparum. Infect. Immun. 59 (10), 3498. Hall, R., Hyde, J.E., Goman, M., Simmons, D.L., Hope, I.A., Mackay, M., Scaife, J., Merkli, B., Richle, R., Stocker, J., 1984. Major surface antigen gene of a human malaria parasite cloned and expressed in bacteria. Nature 311, 379382.

A. Helg et al. / Journal of Immunological Methods 276 (2003) 19–31 Holder, A.A., Freeman, R.R., 1984. The three major antigens on the surface of Plasmodium falciparum merozoites are derived from a single high molecular weight precursor. J. Exp. Med. 160, 624629. Hui, G.S., Siddiqui, W.A., 1987. Serum from Pf195 protected Aotus monkeys inhibit Plasmodium falciparum growth in vitro. Exp. Parasitol. 64, 519522. Jiang, G., Daubenberger, C., Huber, W., Matile, H., Tanner, M., Pluschke, G., 2000. Sequence diversity of the merozoite surface protein 1 of Plasmodium falciparum in clinical isolates from the Kilombero District, Tanzania. Acta Trop. 74 (1), 51. Kumar, S., Collins, W., Egan, A., Yadava, A., Garraud, O., Blackman, M.J., Guevara Patino, J.A., Diggs, C., Kaslow, D.C., 2000. Immunogenicity and efficacy in aotus monkeys of four recombinant Plasmodium falciparum vaccines in multiple adjuvant formulations based on the 19-kilodalton C terminus of merozoite surface protein 1. Infect. Immun. 68 (4), 2215. Lefranc, M.P., 2001. IMGT, the international ImMunoGeneTics database. Nucleic Acids Res. 29 (1), 207. Locher, C.P., Tam, L.Q., Chang, S.P., McBride, J.S., Siddiqui, W.A., 1996. Plasmodium falciparum: gp195 tripeptide repeatspecific monoclonal antibody inhibits parasite growth in vitro. Exp. Parasitol. 84 (1), 74. Martinez, C., Lefranc, M.P., 1998. The mouse (Mus musculus) immunoglobulin kappa variable (IGKV) genes and joining (IGKJ) segments. Exp. Clin. Immunogenet. 15 (3), 184. Matile, H., Pink, J.R.L., 1990. Plasmodium falciparum malaria parasite cultures and their use in immunology. In: Lefkovits, I., Benvenuto, P. (Eds.), Immunological Methods. Academic Press, San Diego, pp. 221 – 234. Mattes, M.J., 1997. Binding parameters of antibodies reacting with multivalent antigens: functional affinity or pseudo-affinity. J. Immunol. Methods 202, 97. Miller, L.H., Roberts, T., Shahabuddin, M., McCutchan, T.F., 1993. Analysis of sequence diversity in the Plasmodium falciparum merozoite surface protein-1 (MSP-1). Mol. Biochem. Parasitol. 59, 1. Neri, D., Montigiani, S., Kirkham, P.M., 1996. Biophysical methods for the determination of antibody – antigen affinities. Tibtech 14, 465. Nygren, H., Stenberg, M., 1989. Immunochemistry at interfaces. J. Immunol. Methods 66, 321. Nygren, H., Werthen, M., Stenberg, M., 1987. Kinetics of antibody binding to solid-phase-immobilised antigen. Effect of diffusion rate limitation and steric interaction. J. Immunol. Methods 101 (1), 63.

31

Patarroyo, M.E., Romero, P., Torres, M.L., Clavijo, P., Moreno, A., Martinez, A., Rodriguez, R., Guzman, F., Cabezas, E., 1987. Induction of protective immunity against experimental infection with malaria using synthetic peptides. Nature 328, 629. Patarroyo, M.E., Amador, R., Clavijo, P., Moreno, A., Guzman, F., Romero, P., Tascon, R., Franco, A., Murillo, L.A., Ponton, G., Trujillo, G., 1988. A synthetic vaccine protects humans against challenge with asexual blood stages of Plasmodium falciparum malaria. Nature 332, 158. Pierce, M.M., Raman, C.S., Nall, B.T., 1999. Isothermal titration calorimetry of protein – protein interactions. Methods 19 (2), 213221 (Oct.). Pluschke, G., Joss, A., Marfurt, J., Daubenberger, C., Kashala, O., Zwickl, M., Stief, A., Sansig, G., Schlapfer, B., Linkert, S., van der Putten, H., Hardman, N., Schro¨der, M., 1998. Generation of chimeric monoclonal antibodies from mice that carry human immunoglobulin Cgamma1 heavy of Ckappa light chain gene segments. J. Immunol. Methods 215 (1 – 2), 2737. Richie, T.L., Saul, A., 2002. Progress and challenges for malaria vaccines. Nature 415, 694701. Savoca, R., Schwab, C., Bosshard, H.R., 1991. Epitope mapping employing immobilized synthetic peptides. How specific is the reactivity of these peptides wit the antiserum raised against the parent protein? J. Immunol. Methods 141, 245. Shi, Y.P., Hasnain, S.E., Sacci, J.B., Holloway, B.P., Fujioka, H., Kumar, N., Wohlhueter, R., Hoffman, S.L., Collins, W.E., Lal, A.A., 1999. Immunogenicity and in vitro protective efficacy of a recombinant multistage Plasmodium falciparum candidate vaccine. Proc. Natl. Acad. Sci. U. S. A. 96, 16151620. Siddiqui, W.A., Tam, L.Q., Kramer, K.J., Hui, G.S., Case, S.E., Yamaga, K.M., Chang, S.P., Chan, E.B., Kann, S.C., 1987. Merozoite surface coat precursor protein completely protects Aotus monkeys against Plasmodium falciparum malaria. Proc. Natl. Acad. Sci. U. S. A. 84 (9), 3014. Stenberg, M., Nygren, H., 1989. Kinetics of antigen-antibody reactions at solid-liquid interfaces. J. Immunol. Methods 113 (1), 3. Underwood, P.A., 1993. Problems and pitfalls with measurement of antibody affinity using solid phase binding in the ELISA. J. Immunol. Methods 164, 119. Wiseman, T., Williston, S., Brandt, J.F., Lung-Nan, L., 1989. Rapid measurement of binding constants and heats of binding using a new titration calorimeter. Anal. Biochem. 179, 131.