The Most Polymorphic Residue on Plasmodium falciparum Apical Membrane Antigen 1 Determines Binding of an Invasion-Inhibitory Antibody
http://www.100md.com
感染与免疫杂志 2006年第5期
Department of Biochemistry
CRC for Diagnostics CRC for Vaccine Technology, La Trobe University, Victoria 3086, Australia
University of Maryland School of Pharmacy, 20 Penn Street, Baltimore, Maryland 21201
ABSTRACT
Apical membrane antigen 1 (AMA1) is currently one of the leading malarial vaccine candidates. Anti-AMA1 antibodies can inhibit the invasion of erythrocytes by Plasmodium merozoites and prevent the multiplication of blood-stage parasites. Here we describe an anti-AMA1 monoclonal antibody (MAb 1F9) that inhibits the invasion of Plasmodium falciparum parasites in vitro. We show that both reactivity of MAb 1F9 with AMA1 and MAb 1F9-mediated invasion inhibition were strain specific. Site-directed mutagenesis of a fragment of AMA1 displayed on M13 bacteriophage identified a single polymorphic residue in domain I of AMA1 that is critical for MAb 1F9 binding. The identities of all other polymorphic residues investigated in this domain had little effect on the binding of the antibody. Examination of the P. falciparum AMA1 crystal structure localized this residue to a surface-exposed -helix at the apex of the polypeptide. This description of a polymorphic inhibitory epitope on AMA1 adds supporting evidence to the hypothesis that immune pressure is responsible for the polymorphisms seen in this molecule.
INTRODUCTION
Malaria is currently one of the most significant causes of morbidity and mortality in the developing world, with up to 500 million new cases annually, mainly in sub-Saharan African children (37). With the increase in antimalarial drug resistance, there is a growing requirement for an effective vaccine directed towards Plasmodium falciparum, the causative agent of the most lethal form of the disease. There are currently a number of vaccine candidates under development, some of the most promising of which are derived from asexual blood-stage antigens found on the merozoite surface. One of these is apical membrane antigen 1 (AMA1). AMA1 undergoes at least three posttransitional proteolytic processing events and has recently been shown to be expressed by sporozoites, the infective stage injected into the human host by an infected anopheline mosquito (20, 36). Animal studies using murine and simian models of the human infection (1, 11, 38) and analyses of the anti-AMA1 antibody response of infected humans (18, 30, 34) have shown that particular anti-AMA1 antibodies inhibit invasion of host erythrocytes by the P. falciparum merozoite and thereby interrupt the multiplication of the parasite in the human host.
The function of AMA1 is unknown, as is the mechanism by which antibodies prevent merozoite invasion, but there is a general consensus that AMA1 plays an important role in the invasion process. All apicomplexa examined to date, including Plasmodium spp., possess a gene encoding an AMA1 polypeptide, indicating a conserved role for this molecule in apicomplexan biology (6, 17). AMA1 has the primary structure of a typical type 1 integral membrane protein with conventional signal and transmembrane sequences and a cytosolic domain (10). The ectodomains of all plasmodial AMA1 molecules sequenced thus far contain a conserved pattern of cysteine residues forming intramolecular disulfide bonds that define three subdomains within the ectodomain. Immunization studies have shown that correct disulfide connectivities are required in order to elicit a protective immune response (1, 18, 29). It is therefore extremely likely that any AMA1-based vaccine would require correct tertiary structure in order to elicit an effective antiplasmodial immune response.
Despite the conserved tertiary structure, sequence polymorphisms exist at more than 60 residue positions in the ectodomain (8, 9, 32, 33). The characterization of these polymorphisms and of their distribution in P. falciparum populations suggests that they have arisen as a result of positive selection (2, 14), most probably exerted by the immune response of the human host. These conclusions from population genetic studies are supported by experimental evidence that the sequence polymorphisms in AMA1 allow parasites to avoid the inhibitory effects of anti-AMA1 antibodies. First, immunization of mice with recombinant Plasmodium chabaudi DS AMA1 conferred almost complete protection against homologous infectious challenge but little protection against heterologous challenge with P. chabaudi 556KA (11). Second, the results of in vitro invasion inhibition assays have demonstrated that rabbit antisera generated by immunization with recombinant AMA1 were strongly inhibitory towards homologous parasites but less inhibitory when tested against either of two heterologous parasite strains (16, 18, 22). Third, naturally acquired human antibodies purified on recombinant, refolded 3D7 AMA1 potently inhibited the invasion of 3D7 merozoites in vitro but were less inhibitory for other strains of P. falciparum (18).
The sequence polymorphisms in the AMA1 ectodomain modify the polypeptide such that it is no longer a target for inhibitory antibodies, but in doing so they must not compromise the overall fitness of the molecule. These competing effects would create clusters of polymorphisms along the polypeptide chain where polymorphic positions represent mutation-tolerant, surface-exposed residues interspersed with non-surface-exposed and/or functionally critical but mutation-intolerant residues. This is the case in P. falciparum AMA1 (8, 32, 33), with polymorphisms particularly clustered in domain I. A majority of the P. falciparum AMA1 sequence polymorphisms described are dimorphic; i.e., there are only two alternative amino acids at a residue position in the primary sequence. The remainder have between three and seven alternative amino acids. The relative contributions of dimorphic and polymorphic residues to the evasion of inhibitory antibodies are unknown, but some evidence indicates that the highly polymorphic sites have a more important role than the dimorphic sites (16, 18, 22).
Mapping the epitopes of monoclonal antibodies (MAbs) is a direct approach to establishing which sequence polymorphisms are important for antibody binding. The inhibitory MAb 4G2dc1, which has been studied extensively (13, 23, 31), reacts with AMA1 from a wide variety of P. falciparum isolates and also with Plasmodium reichenowi AMA1. Site-directed mutagenesis experiments have shown that the 4G2dc1 epitope is contained largely in a conserved but poorly structured loop of domain II (31). 4G2dc1 binding peptides that we selected from a random peptide library induced antibodies that inhibited invasion of the 3D7 and HB3 strains of P. falciparum (5). These peptides have no obvious homology with AMA1 but are likely to mimic the topology of the inhibitory epitope on the surface of AMA1. These studies and others with polyclonal antibodies show that conserved regions of AMA1 are capable of inducing an effective anti-AMA1 antibody response despite the importance of polymorphic regions previously discussed.
In previous work we have used phage display of AMA1 fragments as a convenient means of mapping epitopes of both polyclonal and monoclonal antibodies. This technique showed the binding of human antibodies to AMA1 domain III to be particularly dependent on the C490-C507 disulfide bond and resolved the cysteine connectivities in this region of AMA1 that were not clear from the nuclear magnetic resonance data (29). We have also mapped the epitopes of two MAbs (5G8 and 1F9) to the prodomain and domain I, respectively, using random fragments of AMA1 displayed on phage (7); MAb 5G8 recognized a linear epitope N terminal to the first conserved cysteine, whereas MAb 1F9 recognized a conformational epitope requiring a single disulfide bond (C217-C247) to be in place. In the present study we report that the anti-AMA1 MAb 1F9 is a potent inhibitor of merozoite invasion in vitro. We show that the highly polymorphic region of AMA1 N terminal to C217 is essential for MAb 1F9 binding and that single amino acid substitutions at the most highly polymorphic site in AMA1 (residue 197) abolish MAb 1F9 binding.
MATERIALS AND METHODS
Antibodies. The generation and characterization of the anti-AMA1 antibodies used in this study have been described previously (7, 16).
Parasite culture. P. falciparum strains were cultured as previously described and were synchronized by the sorbitol method (25, 39).
Immunoblotting. Synchronized schizont-stage parasites (5 x 106) were purified from a majority of the uninfected erythrocytes by the Percoll purification method (24), extracted by washing in 0.03% saponin, and solubilized in 50 ml sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) sample buffer. The solubilized parasites were then incubated at 100°C for 10 min, separated using 4 to 20% NuPAGE (Invitrogen) Bis-Tris SDS-PAGE under nonreducing conditions, and electroblotted onto a polyvinylidene difluoride membrane. The membrane was blocked with 10% skim milk in phosphate-buffered saline (PBS) and developed with 2 μg ml–1 of either MAb 1F9, 2C5, or 5G8 or a pool of rabbit anti-3D7, -D10, -HB3, and -W2mef AMA1 antisera in PBS, followed by 1μg ml–1 peroxidase-conjugated rabbit anti-mouse or goat anti-rabbit immunoglobulin G (Chemicon) accordingly. The blots were then visualized by ECL (Amersham).
ELISAs. For assays involving binding of recombinant refolded AMA1 to MAbs 1F9, 2C5, and 5G8, 100 μl AMA1 (1 μg ml–1) in coating buffer (0.1 M sodium carbonate, pH 9.5) was applied to Maxisorp (Nunc) enzyme-linked immunosorbent assay (ELISA) plate wells overnight at 4°C. The plate was then washed in carbonate coating buffer and blocked in 10% (wt/vol) skim milk in coating buffer. The plate was washed three times in PBS-Tween (0.1%, vol/vol). Dilutions of MAbs (100 μl) were then added to the wells, incubated for 4 h at room temperature, and washed three times. One hundred microliters of peroxidase-conjugated sheep anti-mouse antibodies (Chemicon) was then added to each well, incubated at room temperature for 1 h, washed four times, and developed with 100 μl H2O2/tetramethylbenzidine substrate (Sigma). The reaction was stopped by the addition of 50 μl 2 N HCl, and then the absorbance was read at 450 nm.
For assays involving binding of phage-displayed AMA1 fragments to either MAb 1F9 or 9E10, 5 μg ml–1 antibody in coating buffer was added to the plate and allowed to adsorb to the plastic overnight at 4°C. The remainder of the ELISA was performed as described above with the following modifications. AMA1-displaying phage preparations were diluted in PBS-Tween and 100 μl added to wells of the ELISA plate. The plate was incubated for 1 h at room temperature and washed four times with PBS-Tween. Peroxidase-conjugated anti-M13 MAb (Amersham) was then added to each well and incubated, and the ELISA was developed as discussed above. All assays were carried out using duplicate wells.
Invasion inhibition. Invasion inhibition assays were carried out as previously described (4, 22). Briefly, antibody preparations were extensively dialyzed against human tonicity PBS and diluted to four times their final assay concentration in human tonicity PBS. Fifty milliliters of antibody dilution was placed in three replicate wells of a 96-well flat-bottom flat-well tissue culture plate (Nunc). Fifty microliters of complete culture medium was added to the wells, and then 50 ml of mature, synchronized parasites was added to each well to give a 0.3% parasitemia and 2% hematocrit. Uninfected red blood cell (RBC) controls and controls without antibody were included. The assay mixture was incubated for 40 h at 37°C in a moist atmosphere of 94% N2, 1% O2, and 5% CO2. The assay was then developed as previously described (4, 22), and the percent inhibition was calculated using the following formula: 100% – {[(A650 test antibody – A650 RBC)/(A650 negative control – A650 RBC)] x 100%}.
Cloning of AMA1 fragments and domain I into phagemid vector pHENH6. Oligonucleotide primers were designed in order to clone 3D7 AMA1 domain I and fragments thereof into the phagemid vector pHENH6 via PstI and NotI restriction sites (the cloning strategy is illustrated in Fig. 1.). PCR was carried out using P. falciparum 3D7 genomic DNA as the template. PCR products and pHENH6 vector were restriction digested, and ligations were carried out using Ready-to-Go T4 ligase (Pharmacia). After purification, the ligation products were used to transform electrocompetent TG1 Escherichia coli, followed by overnight culture on 2x yeast extract-tryptone (YT) agar containing 50 μg ml–1 ampicillin. The resulting colonies were screened by PCR for the presence of an insert of the correct size. These PCR products were sequenced to confirm correct sequence integrity, orientation, and frame.
Preparation of phage clones. TG1 clones containing the pHENH6/AMA1 fragments were grown to an optical density of 2.0 in 10 ml of 2x TY broth with 50 mg ml–1 ampicillin. M13K07 helper phage (1 x 1012 PFU) was added and allowed to infect, and then 10 ml culture was added to 200 ml "superbroth" containing 70 μg ml–1 kanamycin and 50 μg ml–1 ampicillin and incubated at 37°C for 16 h with shaking to allow for phage production. The phage preparation was centrifuged at 8,000 x g for 15 min to pellet the bacteria, and 50 ml 5x polyethylene glycol-NaCl solution was added to the supernatant containing the phage. The phage-polyethylene glycol was incubated on ice for 4 h to allow the phage to precipitate. Following phage precipitation, the preparation was centrifuged at 8,000 x g for 15 min in order to pellet the phage. The supernatant was discarded, and the phage was resuspended in 1 ml PBS and stored at –70°C. Phage clones were normalized for insert concentrations using the c-Myc epitope tag immediately downstream from the NotI site in the pHENH6 vector. An ELISA was performed using MAb 9E10 (anti-Myc) on serial dilutions of individual phage preparations displaying the various AMA1 fragments to be analyzed. The phage preparations were then diluted such that they all produced similar titration curves when the ELISA was repeated on those dilutions. At this point the preparations all possessed similar levels of Myc epitope tag and therefore equivalent levels of the respective AMA1 fragments.
Generation of phage-displayed AMA1 mutants. Phage displaying the 3D7 AMA1 domain I were cultivated in E. coli CJ236 in the presence of uridine to yield phage particles containing single-stranded, uracilated DNA genomes, and single-stranded DNA was purified (QIAGEN). Mutants were generated using the Kunkel method (35). Briefly, mutagenizing oligonucleotides were designed, phosphorylated, and allowed to anneal to the previously purified single-stranded DNA. Double-stranded DNA was generated by the addition of T7 polymerase, deoxynucleoside triphosphates, and T4 ligase. After double-stranded DNA generation and purification, electrocompetent E. coli TG1 was transformed and propagated on 2x TY agar containing 50 μg ml–1 ampicillin overnight at 37°C. PCR analysis was carried out on the resulting colonies using oligonucleotide primers designed to amplify the insert with a short sequence of flanking vector. The PCR products were sequenced in order to confirm the presence of the mutation. Phage preparations of the mutant AMA1 domains I were then propagated as described above.
RESULTS
MAbs 1F9 and 2C5 are reactive with P. falciparum AMA1 in a strain-specific manner. The specificities of the anti-P. falciparum AMA1 monoclonal antibodies were analyzed by ELISA (Fig. 2A) and immunoblotting (Fig. 2B) using recombinant and parasite-derived sources of AMA1. MAb 1F9 bound to recombinant refolded 3D7- and D10-derived AMA1 but showed very little reactivity with recombinant AMA1 derived from HB3 or W2mef. MAb 1F9 recognized both the 83-kDa unprocessed and 66-kDa processed forms of parasite-derived AMA1 from 3D7 and D10 but failed to recognize AMA1 expressed by HB3 or W2mef parasites (Fig. 2B). This was consistent with the ELISA data and strongly suggests that the lack of reactivity is due to the sequence differences in domain I between 3D7 and the latter two strains. The MAb 2C5 reaction pattern was identical to that of MAb 1F9 in both ELISA and immunoblots. MAb 2C5 requires the entire correctly folded ectodomain for reactivity (J. Schloegel, personal communication) and therefore must recognize an epitope distinct from that of MAb 1F9 but common to both the 3D7 and D10 strains. In contrast to MAbs 1F9 and 2C5, MAb 5G8 reacted equivalently with all four recombinant AMA1s (Fig. 2A) and therefore acted as a convenient control to standardize the level of AMA1 bound to the assay plate. In immunoblots of parasite-derived material, this antibody recognized only the unprocessed (83-kDa) form of AMA1 from all four strains, consistent with the presence of the A-Y-P epitope motif (7) present in the prosequence of AMA1 from every strain of P. falciparum. The pooled rabbit anti-AMA1 reagent recognized the unprocessed (83-kDa) and processed (66-kDa) forms of AMA1 from all strains of parasite tested (Fig. 2B), indicating approximately equivalent levels of AMA1 in each lane.
MAb 1F9 but not MAb 2C5 inhibits the invasion of erythrocytes by 3D7 and D10 merozoites. It is clear that MAb 1F9 is a potent inhibitor of invasion of D10 and 3D7 parasites but not of HB3 or W2mef (Fig. 3A). MAb 2C5 exhibited little or no inhibition at the same antibody concentrations (Fig. 3B). Rabbit anti-AMA1 immunoglobulin G (18) used in the same assay inhibited invasion with significantly higher potency than MAb 1F9 (Fig. 3C).
Defining the "minimal" antigen requirement for 1F9 reactivity. In an attempt to define the minimal requirements for reactivity with the inhibitory MAb 1F9, a fragment of AMA1 domain I spanning residues 179 to 247 was expressed on phage. This 69-residue fragment contains the MAb 1F9 binding region previously reported (7). Phage-expressing fragments progressively truncated from the N terminus were also produced (see Fig. 1B for relative positions in domain I). MAb 1F9 was reactive with the full-length fragment (residues 179 to 247) and with the fragment truncated by 12 residues (residues 191 to 247) but not with any of the other truncated fragments (residues 199 to 247, 204 to 247, and 209 to 247) (Fig. 4). All fragments were expressed equally as judged by their reactivity with the antibody to the Myc tag (29).
Amino acid polymorphisms within AMA1 are responsible for loss of reactivity with MAb 1F9. The fragment of AMA1 recognized by MAb 1F9 is the most polymorphic region of AMA1 and includes residue 197, the most polymorphic site in AMA1 (see Fig. 1A for sequence information). Because the loss of a seven-residue sequence (residues 191 to 198), which included residue 197, ablated MAb 1F9 binding, polymorphism scanning was used to assess the significance of this residue for MAb 1F9 binding. Phage displaying wild-type and mutated forms of P. falciparum 3D7 AMA1 domain I were assayed for reactivity with MAb 1F9. Mutation of residue 197 (E in 3D7) to D (as found in W2mef), H, G, or R completely abolished binding of MAb 1F9 (Fig. 5). Mutating this residue to either Q (as in HB3) or V resulted in significantly reduced reactivity. These results are consistent with the earlier finding when MAb 1F9 reactivity was assayed by ELISA on the recombinant ectodomains derived from 3D7, HB3, and W2mef (Fig. 2A). All of the other polymorphic substitutions at positions 196, 230, 242, 243, and 244 had no effect on MAb 1F9 binding, with the exception that 242 D replaced with Y consistently gave a slightly higher reactivity. All 196/197 double mutants failed to react with MAb 1F9. This presumably reflects the dominant effect of mutations at residue 197. These data indicate that of the AMA1 polymorphic residues examined here (residues 196, 197, 230, 242, 243, and 244), only residue 197 significantly contributes to the binding energy of the inhibitory MAb 1F9.
DISCUSSION
In this study we describe two MAbs (1F9 and 2C5) that bind P. falciparum AMA1 in a conformation-dependent manner, but only one (1F9) is capable of inhibiting merozoite invasion in vitro. Furthermore, we have demonstrated that the binding of MAb1F9 can be prevented by mutations at residue 197, the most polymorphic site in AMA1. This result provides evidence supporting the hypothesis that the cluster of sequence polymorphisms in domain I of AMA1 have been selected for by protective antibody responses (16, 18, 32, 33).
Because conformational epitopes in AMA1 are important for protection against malaria (1), in this study we sought to determine the fine structure of a conformation-dependent inhibitory epitope of AMA1 and to assess the impact of naturally occurring sequence polymorphisms within this epitope on antibody binding. Of the four strains of P. falciparum tested here, only the invasion of 3D7 and D10 was inhibited by MAb 1F9. The lack of sensitivity of HB3 and W2mef parasites to inhibition by MAb 1F9 correlates with the lack of reactivity seen in the ELISA and immunoblot experiments and confirms the strain specificity of this antibody. The sequences of 3D7 and D10 AMA1 are identical in domain I but differ in domains II and III. As we have localized the MAb 1F9 binding site to domain I, it is perhaps not surprising that these parasite strains were both inhibited by this MAb. MAb 2C5, which recognizes a complex conformational epitope in 3D7 and D10 AMA1 requiring the presence of domains I, II, and III (data not shown), did not inhibit invasion. This indicates that antibody binding to conformational epitopes per se is not sufficient to confer invasion inhibition. The number and location of inhibitory epitopes on the surface of AMA1 remain to be determined, but it is clear from this and previous studies that domains I, II, and III are all targets of inhibitory antibodies (7, 29, 31).
The greater potency of the polyclonal anti-AMA1 preparations compared to MAb 1F9 (Fig. 3.) might reflect the additive or even synergistic effect of antibodies recognizing multiple distinct epitopes on the surface of AMA1. Such an effect has been reported in antibody-mediated neutralization of human immunodeficiency virus (27).
MAb 1F9 was initially described in our earlier study (7) in which we identified a 57-residue fragment of AMA1 domain I comprising residues 191 to 247, including the disulfide-bonded C217 and C247. This fragment was necessary and sufficient for MAb 1F9 binding. Shorter fragments generated by successive truncations from the N terminus of the 57-residue fragment did not retain MAb 1F9 binding. Therefore, either contact residues for MAb 1F9 reside in the eight-amino-acid sequence between residues 191 and 199 or deletion of these residues prevented this fragment of AMA1 from adopting the native fold. Two polymorphic residues reside within this eight-residue sequence. Residue 196 is strictly dimorphic, but 197 is the most polymorphic residue in the AMA1 ectodomain, with eight different amino acids described at this position. The results of population genetic studies have indicated that residue 197 may be important in immune evasion (32, 33), and here we have established that residue 197 is important for the binding of an inhibitory, strain-specific MAb, MAb 1F9. Several mutations were created in the context of 3D7 AMA1 domain I; these substitutions in AMA1 were selected to reflect the gene sequences identified in parasite isolates from Africa and Papua New Guinea (8, 9, 32, 33). Since these mutations are found in parasites in the field, the possibility that the overall structure of AMA1 will be disrupted due to the introduction of the mutation is significantly reduced. Of the mutations generated in this study, only those at position 197 resulted in a reduced affinity of MAb 1F9 for AMA1 domain I. With the exception of position 197, no polymorphic residues studied contribute to the MAb 1F9 epitope, as substitution at these positions had little effect on MAb 1F9 binding. One caveat to this was the 242 position, where the replacement of D with Y consistently resulted in an increase in MAb 1F9 binding. Taken together, the experimental information localizes the MAb 1F9 epitope to a short fragment of AMA1 domain I. Five of the six polymorphic residues mutated in this study are located at one end of the fragment, and the sixth residue is located at the other end (Fig. 6A and B). Although residue 197 is a critical component of the epitope, other residues within this fragment must be involved in the overall MAb 1F9 epitope. Further work is under way to identify these residues.
Recently Bai et al. (3) have solved the three-dimensional structure of the domain I+II fragment of P. falciparum AMA1. The 57-residue domain I fragment containing the MAb 1F9 epitope contains the -helix and strands 3 and 4 of the -sheet from the PAN domain of AMA1 domain I identified in the crystal structure (Fig. 6.). Residue 197 is located on a surface-exposed turn of the -helix. The polymorphic residues 196, 242, 243, and 244 that were shown here not to contribute to the 1F9 epitope are in close proximity to residue 197 on the surface of AMA1. However, the structure shows that other polymorphic residues are also in close proximity to residue 197, and it will be a priority to examine the importance of these residues in future studies.
Of the two MAbs used in this study that recognize conformational epitopes on AMA1, one (1F9) was inhibitory whereas the other (2C5) was noninhibitory. The extensively studied 4G2dc1 inhibitory MAb (7, 13, 31) also recognizes a conformationally dependent epitope. Although we have identified residue 197 as a major component of the inhibitory epitope defined by MAb 1F9, until recently a detailed molecular description of the MAb 4G2 epitope has not been available. The broad reactivity of MAb 4G2 with AMA1 molecules from a variety of P. falciparum strains suggests that nonpolymorphic residues contribute substantially to the epitope. Pizarro and colleagues (31) carried out alanine replacements for several residues within P. falciparum AMA1. The residues shown to affect 4G2dc1 binding were identified as nonpolymorphic, and most were localized to a conserved region of domain II in P. falciparum AMA1. It has been previously noted that AMA1 has a "polymorphic" face and a "nonpolymorphic" face, and the positions of the 1F9 and 4G2dcl epitopes on these two faces of the protein are shown in Fig. 6C. Although they are on opposite sides of AMA1, both epitopes flank a deep hydrophobic cleft on the surface of the molecule. It is possible that the interaction of a ligand with this hydrophobic cleft is an important step in the invasion pathway and that binding of either MAb 1F9 or 4G2dc1 could partially obscure this cleft and therefore prevent this interaction. This possible explanation of the invasion-inhibitory properties of these two very different MAbs is currently under investigation in our laboratory.
MAb 1F9 is one of several reagents, including peptides, small proteins, and antibodies, that inhibit invasion and bind to AMA1 (5, 7, 15, 21, 26, 30). Several of these other reagents compete with MAb 1F9 for binding to AMA1 and are therefore likely to bind to the same area of the protein in order to exert their inhibitory effects. Therefore, it is possible that the fragment described in this study is part of a functional "hot spot" on the surface of AMA1, as previously described. It is conceivable that this hot spot is a component of the hydrophobic cleft and may represent an attractive target for small-molecule antimalarial therapy. The identity of residue 197 is clearly important for MAb 1F9 binding, but its importance in a naturally acquired, protective antibody response to AMA1 is unclear. Nevertheless, it would be desirable that an antibody response by an AMA1 vaccine not be dominated by specificities recognizing the MAb 1F9 epitope. With this in mind, consideration should be given to the nature of this and other polymorphic residues in the design and formulation of an AMA1-based vaccine in order to provide protection against a wider cohort of AMA1 genotypes.
The approach taken in this study was to analyze the effect of single polymorphisms introduced into the 3D7 AMA1 domain I on binding of MAb 1F9 when displayed on the surface of M13 bacteriophage. Phage display has been used successfully to identify epitopes in a variety of systems (7, 12, 28), and the technique relies on the faithful display of both continuous and discontinuous epitopes within the proteins displayed on M13 phage. M13 phage display has been shown to allow the formation of even complex epitopes due to the periplasmic folding of proteins during phage production. Previously we have described the display of the complete AMA1 ectodomain in addition to the individual domains on phage and have shown their utility in epitope mapping and structural analyses (7, 29). In this study we have exploited the technology for functional analysis of individual polymorphisms within AMA1. We believe the approach of polymorphism-scanning mutagenesis is a suitable method for determining the effects of individual or multiple polymorphisms on the binding of antibodies to their respective antigens. It obviates the often expensive and time-consuming need to produce multiple recombinant proteins or protein domains, some of which require complex folding and disulfide bonding in order to represent the authentic antigen. Polymorphism scanning also offers a relatively high-throughput procedure for the study of polymorphic residues in vaccine candidate molecules such as P. falciparum AMA1. It is likely that the approach taken here is highly suitable for the analysis of the immunological significance of polymorphisms in other protein antigens, both in malaria parasites and in other pathogens.
ACKNOWLEDGMENTS
This work was supported in part by the National Health and Medical Research Council of Australia and the National Institutes of Health (NIH grant R01AI59229).
REFERENCES
1. Anders, R. F., P. E. Crewther, S. Edwards, M. Margetts, M. L. Matthew, B. Pollock, and D. Pye. 1998. Immunisation with recombinant AMA-1 protects mice against infection with Plasmodium chabaudi. Vaccine 16:240-247.
2. Anders, R. F., D. J. McColl, and R. L. Coppel. 1993. Molecular variation in Plasmodium falciparum: polymorphic antigens of asexual erythrocytic stages. Acta Trop. 53:239-253.
3. Bai, T., M. Becker, A. Gupta, P. Strike, V. J. Murphy, R. F. Anders, and A. H. Batchelor. 2005. Structure of AMA1 from Plasmodium falciparum reveals a clustering of polymorphisms that surround a conserved hydrophobic pocket. Proc. Natl. Acad. Sci. USA 102:12736-12741.
4. Basco, L. K., F. Marquet, M. M. Makler, and J. Le Bras. 1995. Plasmodium falciparum and Plasmodium vivax: lactate dehydrogenase activity and its application for in vitro drug susceptibility assay. Exp. Parasitol. 80:260-271.
5. Casey, J. L., A. M. Coley, R. F. Anders, V. J. Murphy, K. S. Humberstone, A. W. Thomas, and M. Foley. 2004. Antibodies to malaria peptide mimics inhibit Plasmodium falciparum invasion of erythrocytes. Infect. Immun. 72:1126-1134.
6. Chesne-Seck, M. L., J. C. Pizarro, B. V. Normand, C. R. Collins, M. J. Blackman, B. W. Faber, E. J. Remarque, C. H. Kocken, A. W. Thomas, and G. A. Bentley. 2005. Structural comparison of apical membrane antigen 1 orthologues and paralogues in apicomplexan parasites. Mol. Biochem. Parasitol. 144:55-67.
7. Coley, A. M., N. V. Campanale, J. L. Casey, A. N. Hodder, P. E. Crewther, R. F. Anders, L. M. Tilley, and M. Foley. 2001. Rapid and precise epitope mapping of monoclonal antibodies against Plasmodium falciparum AMA1 by combined phage display of fragments and random peptides. Protein Eng. 14:691-698.
8. Cortes, A., M. Mellombo, R. Masciantonio, V. J. Murphy, J. C. Reeder, and R. F. Anders. 2005. Allele specificity of naturally acquired antibody responses against Plasmodium falciparum apical membrane antigen 1. Infect. Immun. 73:422-430.
9. Cortes, A., M. Mellombo, I. Mueller, A. Benet, J. C. Reeder, and R. F. Anders. 2003. Geographical structure of diversity and differences between symptomatic and asymptomatic infections for Plasmodium falciparum vaccine candidate AMA1. Infect. Immun. 71:1416-1426.
10. Crewther, P. E., J. G. Culvenor, A. Silva, J. A. Cooper, and R. F. Anders. 1990. Plasmodium falciparum: two antigens of similar size are located in different compartments of the rhoptry. Exp. Parasitol. 70:193-206.
11. Crewther, P. E., M. L. Matthew, R. H. Flegg, and R. F. Anders. 1996. Protective immune responses to apical membrane antigen 1 of Plasmodium chabaudi involve recognition of strain-specific epitopes. Infect. Immun. 64:3310-3317.
12. Cui, X., H. S. Nagesha, and I. H. Holmes. 2003. Mapping of conformational epitopes on capsid protein VP2 of infectious bursal disease virus by fd-tet phage display. J. Virol. Methods 114:109-112.
13. Dutta, S., J. D. Haynes, A. Barbosa, L. A. Ware, J. D. Snavely, J. K. Moch, A. W. Thomas, and D. E. Lanar. 2005. Mode of action of invasion-inhibitory antibodies directed against apical membrane antigen 1 of Plasmodium falciparum. Infect. Immun. 73:2116-2122.
14. Escalante, A. A., A. A. Lal, and F. J. Ayala. 1998. Genetic polymorphism and natural selection in the malaria parasite Plasmodium falciparum. Genetics 149:189-202.
15. Harris, K. S., J. L. Casey, A. M. Coley, R. Masciantonio, J. K. Sabo, D. W. Keizer, E. F. Lee, A. McMahon, R. S. Norton, R. F. Anders, and M. Foley. 2005. Binding hot spot for invasion inhibitory molecules on Plasmodium falciparum apical membrane antigen 1. Infect. Immun. 73:6981-6989.
16. Healer, J., V. Murphy, A. N. Hodder, R. Masciantonio, A. W. Gemmill, R. F. Anders, A. F. Cowman, and A. Batchelor. 2004. Allelic polymorphisms in apical membrane antigen-1 are responsible for evasion of antibody-mediated inhibition in Plasmodium falciparum. Mol. Microbiol. 52:159-168.
17. Hehl, A. B., C. Lekutis, M. E. Grigg, P. J. Bradley, J. F. Dubremetz, E. Ortega-Barria, and J. C. Boothroyd. 2000. Toxoplasma gondii homologue of plasmodium apical membrane antigen 1 is involved in invasion of host cells. Infect. Immun. 68:7078-7086.
18. Hodder, A. N., P. E. Crewther, and R. F. Anders. 2001. Specificity of the protective antibody response to apical membrane antigen 1. Infect. Immun. 69:3286-3294.
19. Hodder, A. N., P. E. Crewther, M. L. Matthew, G. E. Reid, R. L. Moritz, R. J. Simpson, and R. F. Anders. 1996. The disulfide bond structure of Plasmodium apical membrane antigen-1. J. Biol. Chem. 271:29446-29452.
20. Howell, S. A., F. Hackett, A. M. Jongco, C. Withers-Martinez, K. Kim, V. B. Carruthers, and M. J. Blackman. 2005. Distinct mechanisms govern proteolytic shedding of a key invasion protein in apicomplexan pathogens. Mol. Microbiol. 57:1342-1356.
21. Keizer, D. W., L. A. Miles, F. Li, M. Nair, R. F. Anders, A. M. Coley, M. Foley, and R. S. Norton. 2003. Structures of phage-display peptides that bind to the malarial surface protein, apical membrane antigen 1, and block erythrocyte invasion. Biochemistry 42:9915-9923.
22. Kennedy, M. C., J. Wang, Y. Zhang, A. P. Miles, F. Chitsaz, A. Saul, C. A. Long, L. H. Miller, and A. W. Stowers. 2002. In vitro studies with recombinant Plasmodium falciparum apical membrane antigen 1 (AMA1): production and activity of an AMA1 vaccine and generation of a multiallelic response. Infect. Immun. 70:6948-6960.
23. Kocken, C. H., D. L. Narum, A. Massougbodji, B. Ayivi, M. A. Dubbeld, A. van der Wel, D. J. Conway, A. Sanni, and A. W. Thomas. 2000. Molecular characterisation of Plasmodium reichenowi apical membrane antigen-1 (AMA-1), comparison with P. falciparum AMA-1, and antibody-mediated inhibition of red cell invasion. Mol. Biochem. Parasitol. 109:147-156.
24. Kramer, K. J., S. C. Kan, and W. A. Siddiqui. 1982. Concentration of Plasmodium falciparum-infected erythrocytes by density gradient centrifugation in Percoll. J. Parasitol. 68:336-337.
25. Lambros, C., and J. P. Vanderberg. 1979. Synchronization of Plasmodium falciparum erythrocytic stages in culture. J. Parasitol. 65:418-420.
26. Li, F., A. Dluzewski, A. M. Coley, A. Thomas, L. Tilley, R. F. Anders, and M. Foley. 2002. Phage-displayed peptides bind to the malarial protein apical membrane antigen-1 and inhibit the merozoite invasion of host erythrocytes. J. Biol. Chem. 277:50303-50310.
27. Mascola, J. R., M. K. Louder, T. C. VanCott, C. V. Sapan, J. S. Lambert, L. R. Muenz, B. Bunow, D. L. Birx, and M. L. Robb. 1997. Potent and synergistic neutralization of human immunodeficiency virus (HIV) type 1 primary isolates by hyperimmune anti-HIV immunoglobulin combined with monoclonal antibodies 2F5 and 2G12. J. Virol. 71:7198-7206.
28. Muhle, C., S. Schulz-Drost, A. V. Khrenov, E. L. Saenko, J. Klinge, and H. Schneider. 2004. Epitope mapping of polyclonal clotting factor VIII-inhibitory antibodies using phage display. Thromb. Haemost. 91:619-625.
29. Nair, M., M. G. Hinds, A. M. Coley, A. N. Hodder, M. Foley, R. F. Anders, and R. S. Norton. 2002. Structure of domain III of the blood-stage malaria vaccine candidate, Plasmodium falciparum apical membrane antigen 1 (AMA1). J. Mol. Biol. 322:741-753.
30. Nuttall, S. D., K. S. Humberstone, U. V. Krishnan, J. A. Carmichael, L. Doughty, M. Hattarki, A. M. Coley, J. L. Casey, R. F. Anders, M. Foley, R. A. Irving, and P. J. Hudson. 2004. Selection and affinity maturation of IgNAR variable domains targeting Plasmodium falciparum AMA1. Proteins 55:187-197.
31. Pizarro, J. C., B. Vulliez-Le Normand, M. L. Chesne-Seck, C. R. Collins, C. Withers-Martinez, F. Hackett, M. J. Blackman, B. W. Faber, E. J. Remarque, C. H. Kocken, A. W. Thomas, and G. A. Bentley. 2005. Crystal structure of the malaria vaccine candidate apical membrane antigen 1. Science 308:408-411.
32. Polley, S. D., W. Chokejindachai, and D. J. Conway. 2003. Allele frequency-based analyses robustly map sequence sites under balancing selection in a malaria vaccine candidate antigen. Genetics 165:555-561.
33. Polley, S. D., and D. J. Conway. 2001. Strong diversifying selection on domains of the Plasmodium falciparum apical membrane antigen 1 gene. Genetics 158:1505-1512.
34. Polley, S. D., T. Mwangi, C. H. Kocken, A. W. Thomas, S. Dutta, D. E. Lanar, E. Remarque, A. Ross, T. N. Williams, G. Mwambingu, B. Lowe, D. J. Conway, and K. Marsh. 2004. Human antibodies to recombinant protein constructs of Plasmodium falciparum apical membrane antigen 1 (AMA1) and their associations with protection from malaria. Vaccine 23:718-728.
35. Sidhu, S. S., H. B. Lowman, B. C. Cunningham, and J. A. Wells. 2000. Phage display for selection of novel binding peptides. Methods Enzymol. 328:333-363.
36. Silvie, O., J. F. Franetich, S. Charrin, M. S. Mueller, A. Siau, M. Bodescot, E. Rubinstein, L. Hannoun, Y. Charoenvit, C. H. Kocken, A. W. Thomas, G. J. Van Gemert, R. W. Sauerwein, M. J. Blackman, R. F. Anders, G. Pluschke, and D. Mazier. 2004. A role for apical membrane antigen 1 during invasion of hepatocytes by Plasmodium falciparum sporozoites. J. Biol. Chem. 279:9490-9496.
37. Snow, R. W., C. A. Guerra, A. M. Noor, H. Y. Myint, and S. I. Hay. 2005. The global distribution of clinical episodes of Plasmodium falciparum malaria. Nature 434:214-217.
38. Stowers, A. W., M. C. Kennedy, B. P. Keegan, A. Saul, C. A. Long, and L. H. Miller. 2002. Vaccination of monkeys with recombinant Plasmodium falciparum apical membrane antigen 1 confers protection against blood-stage malaria. Infect. Immun. 70:6961-6967.
39. Trager, W., and J. B. Jensen. 1976. Human malaria parasites in continuous culture. Science 193:673-675.(A. M. Coley, K. Parisi, R)
CRC for Diagnostics CRC for Vaccine Technology, La Trobe University, Victoria 3086, Australia
University of Maryland School of Pharmacy, 20 Penn Street, Baltimore, Maryland 21201
ABSTRACT
Apical membrane antigen 1 (AMA1) is currently one of the leading malarial vaccine candidates. Anti-AMA1 antibodies can inhibit the invasion of erythrocytes by Plasmodium merozoites and prevent the multiplication of blood-stage parasites. Here we describe an anti-AMA1 monoclonal antibody (MAb 1F9) that inhibits the invasion of Plasmodium falciparum parasites in vitro. We show that both reactivity of MAb 1F9 with AMA1 and MAb 1F9-mediated invasion inhibition were strain specific. Site-directed mutagenesis of a fragment of AMA1 displayed on M13 bacteriophage identified a single polymorphic residue in domain I of AMA1 that is critical for MAb 1F9 binding. The identities of all other polymorphic residues investigated in this domain had little effect on the binding of the antibody. Examination of the P. falciparum AMA1 crystal structure localized this residue to a surface-exposed -helix at the apex of the polypeptide. This description of a polymorphic inhibitory epitope on AMA1 adds supporting evidence to the hypothesis that immune pressure is responsible for the polymorphisms seen in this molecule.
INTRODUCTION
Malaria is currently one of the most significant causes of morbidity and mortality in the developing world, with up to 500 million new cases annually, mainly in sub-Saharan African children (37). With the increase in antimalarial drug resistance, there is a growing requirement for an effective vaccine directed towards Plasmodium falciparum, the causative agent of the most lethal form of the disease. There are currently a number of vaccine candidates under development, some of the most promising of which are derived from asexual blood-stage antigens found on the merozoite surface. One of these is apical membrane antigen 1 (AMA1). AMA1 undergoes at least three posttransitional proteolytic processing events and has recently been shown to be expressed by sporozoites, the infective stage injected into the human host by an infected anopheline mosquito (20, 36). Animal studies using murine and simian models of the human infection (1, 11, 38) and analyses of the anti-AMA1 antibody response of infected humans (18, 30, 34) have shown that particular anti-AMA1 antibodies inhibit invasion of host erythrocytes by the P. falciparum merozoite and thereby interrupt the multiplication of the parasite in the human host.
The function of AMA1 is unknown, as is the mechanism by which antibodies prevent merozoite invasion, but there is a general consensus that AMA1 plays an important role in the invasion process. All apicomplexa examined to date, including Plasmodium spp., possess a gene encoding an AMA1 polypeptide, indicating a conserved role for this molecule in apicomplexan biology (6, 17). AMA1 has the primary structure of a typical type 1 integral membrane protein with conventional signal and transmembrane sequences and a cytosolic domain (10). The ectodomains of all plasmodial AMA1 molecules sequenced thus far contain a conserved pattern of cysteine residues forming intramolecular disulfide bonds that define three subdomains within the ectodomain. Immunization studies have shown that correct disulfide connectivities are required in order to elicit a protective immune response (1, 18, 29). It is therefore extremely likely that any AMA1-based vaccine would require correct tertiary structure in order to elicit an effective antiplasmodial immune response.
Despite the conserved tertiary structure, sequence polymorphisms exist at more than 60 residue positions in the ectodomain (8, 9, 32, 33). The characterization of these polymorphisms and of their distribution in P. falciparum populations suggests that they have arisen as a result of positive selection (2, 14), most probably exerted by the immune response of the human host. These conclusions from population genetic studies are supported by experimental evidence that the sequence polymorphisms in AMA1 allow parasites to avoid the inhibitory effects of anti-AMA1 antibodies. First, immunization of mice with recombinant Plasmodium chabaudi DS AMA1 conferred almost complete protection against homologous infectious challenge but little protection against heterologous challenge with P. chabaudi 556KA (11). Second, the results of in vitro invasion inhibition assays have demonstrated that rabbit antisera generated by immunization with recombinant AMA1 were strongly inhibitory towards homologous parasites but less inhibitory when tested against either of two heterologous parasite strains (16, 18, 22). Third, naturally acquired human antibodies purified on recombinant, refolded 3D7 AMA1 potently inhibited the invasion of 3D7 merozoites in vitro but were less inhibitory for other strains of P. falciparum (18).
The sequence polymorphisms in the AMA1 ectodomain modify the polypeptide such that it is no longer a target for inhibitory antibodies, but in doing so they must not compromise the overall fitness of the molecule. These competing effects would create clusters of polymorphisms along the polypeptide chain where polymorphic positions represent mutation-tolerant, surface-exposed residues interspersed with non-surface-exposed and/or functionally critical but mutation-intolerant residues. This is the case in P. falciparum AMA1 (8, 32, 33), with polymorphisms particularly clustered in domain I. A majority of the P. falciparum AMA1 sequence polymorphisms described are dimorphic; i.e., there are only two alternative amino acids at a residue position in the primary sequence. The remainder have between three and seven alternative amino acids. The relative contributions of dimorphic and polymorphic residues to the evasion of inhibitory antibodies are unknown, but some evidence indicates that the highly polymorphic sites have a more important role than the dimorphic sites (16, 18, 22).
Mapping the epitopes of monoclonal antibodies (MAbs) is a direct approach to establishing which sequence polymorphisms are important for antibody binding. The inhibitory MAb 4G2dc1, which has been studied extensively (13, 23, 31), reacts with AMA1 from a wide variety of P. falciparum isolates and also with Plasmodium reichenowi AMA1. Site-directed mutagenesis experiments have shown that the 4G2dc1 epitope is contained largely in a conserved but poorly structured loop of domain II (31). 4G2dc1 binding peptides that we selected from a random peptide library induced antibodies that inhibited invasion of the 3D7 and HB3 strains of P. falciparum (5). These peptides have no obvious homology with AMA1 but are likely to mimic the topology of the inhibitory epitope on the surface of AMA1. These studies and others with polyclonal antibodies show that conserved regions of AMA1 are capable of inducing an effective anti-AMA1 antibody response despite the importance of polymorphic regions previously discussed.
In previous work we have used phage display of AMA1 fragments as a convenient means of mapping epitopes of both polyclonal and monoclonal antibodies. This technique showed the binding of human antibodies to AMA1 domain III to be particularly dependent on the C490-C507 disulfide bond and resolved the cysteine connectivities in this region of AMA1 that were not clear from the nuclear magnetic resonance data (29). We have also mapped the epitopes of two MAbs (5G8 and 1F9) to the prodomain and domain I, respectively, using random fragments of AMA1 displayed on phage (7); MAb 5G8 recognized a linear epitope N terminal to the first conserved cysteine, whereas MAb 1F9 recognized a conformational epitope requiring a single disulfide bond (C217-C247) to be in place. In the present study we report that the anti-AMA1 MAb 1F9 is a potent inhibitor of merozoite invasion in vitro. We show that the highly polymorphic region of AMA1 N terminal to C217 is essential for MAb 1F9 binding and that single amino acid substitutions at the most highly polymorphic site in AMA1 (residue 197) abolish MAb 1F9 binding.
MATERIALS AND METHODS
Antibodies. The generation and characterization of the anti-AMA1 antibodies used in this study have been described previously (7, 16).
Parasite culture. P. falciparum strains were cultured as previously described and were synchronized by the sorbitol method (25, 39).
Immunoblotting. Synchronized schizont-stage parasites (5 x 106) were purified from a majority of the uninfected erythrocytes by the Percoll purification method (24), extracted by washing in 0.03% saponin, and solubilized in 50 ml sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) sample buffer. The solubilized parasites were then incubated at 100°C for 10 min, separated using 4 to 20% NuPAGE (Invitrogen) Bis-Tris SDS-PAGE under nonreducing conditions, and electroblotted onto a polyvinylidene difluoride membrane. The membrane was blocked with 10% skim milk in phosphate-buffered saline (PBS) and developed with 2 μg ml–1 of either MAb 1F9, 2C5, or 5G8 or a pool of rabbit anti-3D7, -D10, -HB3, and -W2mef AMA1 antisera in PBS, followed by 1μg ml–1 peroxidase-conjugated rabbit anti-mouse or goat anti-rabbit immunoglobulin G (Chemicon) accordingly. The blots were then visualized by ECL (Amersham).
ELISAs. For assays involving binding of recombinant refolded AMA1 to MAbs 1F9, 2C5, and 5G8, 100 μl AMA1 (1 μg ml–1) in coating buffer (0.1 M sodium carbonate, pH 9.5) was applied to Maxisorp (Nunc) enzyme-linked immunosorbent assay (ELISA) plate wells overnight at 4°C. The plate was then washed in carbonate coating buffer and blocked in 10% (wt/vol) skim milk in coating buffer. The plate was washed three times in PBS-Tween (0.1%, vol/vol). Dilutions of MAbs (100 μl) were then added to the wells, incubated for 4 h at room temperature, and washed three times. One hundred microliters of peroxidase-conjugated sheep anti-mouse antibodies (Chemicon) was then added to each well, incubated at room temperature for 1 h, washed four times, and developed with 100 μl H2O2/tetramethylbenzidine substrate (Sigma). The reaction was stopped by the addition of 50 μl 2 N HCl, and then the absorbance was read at 450 nm.
For assays involving binding of phage-displayed AMA1 fragments to either MAb 1F9 or 9E10, 5 μg ml–1 antibody in coating buffer was added to the plate and allowed to adsorb to the plastic overnight at 4°C. The remainder of the ELISA was performed as described above with the following modifications. AMA1-displaying phage preparations were diluted in PBS-Tween and 100 μl added to wells of the ELISA plate. The plate was incubated for 1 h at room temperature and washed four times with PBS-Tween. Peroxidase-conjugated anti-M13 MAb (Amersham) was then added to each well and incubated, and the ELISA was developed as discussed above. All assays were carried out using duplicate wells.
Invasion inhibition. Invasion inhibition assays were carried out as previously described (4, 22). Briefly, antibody preparations were extensively dialyzed against human tonicity PBS and diluted to four times their final assay concentration in human tonicity PBS. Fifty milliliters of antibody dilution was placed in three replicate wells of a 96-well flat-bottom flat-well tissue culture plate (Nunc). Fifty microliters of complete culture medium was added to the wells, and then 50 ml of mature, synchronized parasites was added to each well to give a 0.3% parasitemia and 2% hematocrit. Uninfected red blood cell (RBC) controls and controls without antibody were included. The assay mixture was incubated for 40 h at 37°C in a moist atmosphere of 94% N2, 1% O2, and 5% CO2. The assay was then developed as previously described (4, 22), and the percent inhibition was calculated using the following formula: 100% – {[(A650 test antibody – A650 RBC)/(A650 negative control – A650 RBC)] x 100%}.
Cloning of AMA1 fragments and domain I into phagemid vector pHENH6. Oligonucleotide primers were designed in order to clone 3D7 AMA1 domain I and fragments thereof into the phagemid vector pHENH6 via PstI and NotI restriction sites (the cloning strategy is illustrated in Fig. 1.). PCR was carried out using P. falciparum 3D7 genomic DNA as the template. PCR products and pHENH6 vector were restriction digested, and ligations were carried out using Ready-to-Go T4 ligase (Pharmacia). After purification, the ligation products were used to transform electrocompetent TG1 Escherichia coli, followed by overnight culture on 2x yeast extract-tryptone (YT) agar containing 50 μg ml–1 ampicillin. The resulting colonies were screened by PCR for the presence of an insert of the correct size. These PCR products were sequenced to confirm correct sequence integrity, orientation, and frame.
Preparation of phage clones. TG1 clones containing the pHENH6/AMA1 fragments were grown to an optical density of 2.0 in 10 ml of 2x TY broth with 50 mg ml–1 ampicillin. M13K07 helper phage (1 x 1012 PFU) was added and allowed to infect, and then 10 ml culture was added to 200 ml "superbroth" containing 70 μg ml–1 kanamycin and 50 μg ml–1 ampicillin and incubated at 37°C for 16 h with shaking to allow for phage production. The phage preparation was centrifuged at 8,000 x g for 15 min to pellet the bacteria, and 50 ml 5x polyethylene glycol-NaCl solution was added to the supernatant containing the phage. The phage-polyethylene glycol was incubated on ice for 4 h to allow the phage to precipitate. Following phage precipitation, the preparation was centrifuged at 8,000 x g for 15 min in order to pellet the phage. The supernatant was discarded, and the phage was resuspended in 1 ml PBS and stored at –70°C. Phage clones were normalized for insert concentrations using the c-Myc epitope tag immediately downstream from the NotI site in the pHENH6 vector. An ELISA was performed using MAb 9E10 (anti-Myc) on serial dilutions of individual phage preparations displaying the various AMA1 fragments to be analyzed. The phage preparations were then diluted such that they all produced similar titration curves when the ELISA was repeated on those dilutions. At this point the preparations all possessed similar levels of Myc epitope tag and therefore equivalent levels of the respective AMA1 fragments.
Generation of phage-displayed AMA1 mutants. Phage displaying the 3D7 AMA1 domain I were cultivated in E. coli CJ236 in the presence of uridine to yield phage particles containing single-stranded, uracilated DNA genomes, and single-stranded DNA was purified (QIAGEN). Mutants were generated using the Kunkel method (35). Briefly, mutagenizing oligonucleotides were designed, phosphorylated, and allowed to anneal to the previously purified single-stranded DNA. Double-stranded DNA was generated by the addition of T7 polymerase, deoxynucleoside triphosphates, and T4 ligase. After double-stranded DNA generation and purification, electrocompetent E. coli TG1 was transformed and propagated on 2x TY agar containing 50 μg ml–1 ampicillin overnight at 37°C. PCR analysis was carried out on the resulting colonies using oligonucleotide primers designed to amplify the insert with a short sequence of flanking vector. The PCR products were sequenced in order to confirm the presence of the mutation. Phage preparations of the mutant AMA1 domains I were then propagated as described above.
RESULTS
MAbs 1F9 and 2C5 are reactive with P. falciparum AMA1 in a strain-specific manner. The specificities of the anti-P. falciparum AMA1 monoclonal antibodies were analyzed by ELISA (Fig. 2A) and immunoblotting (Fig. 2B) using recombinant and parasite-derived sources of AMA1. MAb 1F9 bound to recombinant refolded 3D7- and D10-derived AMA1 but showed very little reactivity with recombinant AMA1 derived from HB3 or W2mef. MAb 1F9 recognized both the 83-kDa unprocessed and 66-kDa processed forms of parasite-derived AMA1 from 3D7 and D10 but failed to recognize AMA1 expressed by HB3 or W2mef parasites (Fig. 2B). This was consistent with the ELISA data and strongly suggests that the lack of reactivity is due to the sequence differences in domain I between 3D7 and the latter two strains. The MAb 2C5 reaction pattern was identical to that of MAb 1F9 in both ELISA and immunoblots. MAb 2C5 requires the entire correctly folded ectodomain for reactivity (J. Schloegel, personal communication) and therefore must recognize an epitope distinct from that of MAb 1F9 but common to both the 3D7 and D10 strains. In contrast to MAbs 1F9 and 2C5, MAb 5G8 reacted equivalently with all four recombinant AMA1s (Fig. 2A) and therefore acted as a convenient control to standardize the level of AMA1 bound to the assay plate. In immunoblots of parasite-derived material, this antibody recognized only the unprocessed (83-kDa) form of AMA1 from all four strains, consistent with the presence of the A-Y-P epitope motif (7) present in the prosequence of AMA1 from every strain of P. falciparum. The pooled rabbit anti-AMA1 reagent recognized the unprocessed (83-kDa) and processed (66-kDa) forms of AMA1 from all strains of parasite tested (Fig. 2B), indicating approximately equivalent levels of AMA1 in each lane.
MAb 1F9 but not MAb 2C5 inhibits the invasion of erythrocytes by 3D7 and D10 merozoites. It is clear that MAb 1F9 is a potent inhibitor of invasion of D10 and 3D7 parasites but not of HB3 or W2mef (Fig. 3A). MAb 2C5 exhibited little or no inhibition at the same antibody concentrations (Fig. 3B). Rabbit anti-AMA1 immunoglobulin G (18) used in the same assay inhibited invasion with significantly higher potency than MAb 1F9 (Fig. 3C).
Defining the "minimal" antigen requirement for 1F9 reactivity. In an attempt to define the minimal requirements for reactivity with the inhibitory MAb 1F9, a fragment of AMA1 domain I spanning residues 179 to 247 was expressed on phage. This 69-residue fragment contains the MAb 1F9 binding region previously reported (7). Phage-expressing fragments progressively truncated from the N terminus were also produced (see Fig. 1B for relative positions in domain I). MAb 1F9 was reactive with the full-length fragment (residues 179 to 247) and with the fragment truncated by 12 residues (residues 191 to 247) but not with any of the other truncated fragments (residues 199 to 247, 204 to 247, and 209 to 247) (Fig. 4). All fragments were expressed equally as judged by their reactivity with the antibody to the Myc tag (29).
Amino acid polymorphisms within AMA1 are responsible for loss of reactivity with MAb 1F9. The fragment of AMA1 recognized by MAb 1F9 is the most polymorphic region of AMA1 and includes residue 197, the most polymorphic site in AMA1 (see Fig. 1A for sequence information). Because the loss of a seven-residue sequence (residues 191 to 198), which included residue 197, ablated MAb 1F9 binding, polymorphism scanning was used to assess the significance of this residue for MAb 1F9 binding. Phage displaying wild-type and mutated forms of P. falciparum 3D7 AMA1 domain I were assayed for reactivity with MAb 1F9. Mutation of residue 197 (E in 3D7) to D (as found in W2mef), H, G, or R completely abolished binding of MAb 1F9 (Fig. 5). Mutating this residue to either Q (as in HB3) or V resulted in significantly reduced reactivity. These results are consistent with the earlier finding when MAb 1F9 reactivity was assayed by ELISA on the recombinant ectodomains derived from 3D7, HB3, and W2mef (Fig. 2A). All of the other polymorphic substitutions at positions 196, 230, 242, 243, and 244 had no effect on MAb 1F9 binding, with the exception that 242 D replaced with Y consistently gave a slightly higher reactivity. All 196/197 double mutants failed to react with MAb 1F9. This presumably reflects the dominant effect of mutations at residue 197. These data indicate that of the AMA1 polymorphic residues examined here (residues 196, 197, 230, 242, 243, and 244), only residue 197 significantly contributes to the binding energy of the inhibitory MAb 1F9.
DISCUSSION
In this study we describe two MAbs (1F9 and 2C5) that bind P. falciparum AMA1 in a conformation-dependent manner, but only one (1F9) is capable of inhibiting merozoite invasion in vitro. Furthermore, we have demonstrated that the binding of MAb1F9 can be prevented by mutations at residue 197, the most polymorphic site in AMA1. This result provides evidence supporting the hypothesis that the cluster of sequence polymorphisms in domain I of AMA1 have been selected for by protective antibody responses (16, 18, 32, 33).
Because conformational epitopes in AMA1 are important for protection against malaria (1), in this study we sought to determine the fine structure of a conformation-dependent inhibitory epitope of AMA1 and to assess the impact of naturally occurring sequence polymorphisms within this epitope on antibody binding. Of the four strains of P. falciparum tested here, only the invasion of 3D7 and D10 was inhibited by MAb 1F9. The lack of sensitivity of HB3 and W2mef parasites to inhibition by MAb 1F9 correlates with the lack of reactivity seen in the ELISA and immunoblot experiments and confirms the strain specificity of this antibody. The sequences of 3D7 and D10 AMA1 are identical in domain I but differ in domains II and III. As we have localized the MAb 1F9 binding site to domain I, it is perhaps not surprising that these parasite strains were both inhibited by this MAb. MAb 2C5, which recognizes a complex conformational epitope in 3D7 and D10 AMA1 requiring the presence of domains I, II, and III (data not shown), did not inhibit invasion. This indicates that antibody binding to conformational epitopes per se is not sufficient to confer invasion inhibition. The number and location of inhibitory epitopes on the surface of AMA1 remain to be determined, but it is clear from this and previous studies that domains I, II, and III are all targets of inhibitory antibodies (7, 29, 31).
The greater potency of the polyclonal anti-AMA1 preparations compared to MAb 1F9 (Fig. 3.) might reflect the additive or even synergistic effect of antibodies recognizing multiple distinct epitopes on the surface of AMA1. Such an effect has been reported in antibody-mediated neutralization of human immunodeficiency virus (27).
MAb 1F9 was initially described in our earlier study (7) in which we identified a 57-residue fragment of AMA1 domain I comprising residues 191 to 247, including the disulfide-bonded C217 and C247. This fragment was necessary and sufficient for MAb 1F9 binding. Shorter fragments generated by successive truncations from the N terminus of the 57-residue fragment did not retain MAb 1F9 binding. Therefore, either contact residues for MAb 1F9 reside in the eight-amino-acid sequence between residues 191 and 199 or deletion of these residues prevented this fragment of AMA1 from adopting the native fold. Two polymorphic residues reside within this eight-residue sequence. Residue 196 is strictly dimorphic, but 197 is the most polymorphic residue in the AMA1 ectodomain, with eight different amino acids described at this position. The results of population genetic studies have indicated that residue 197 may be important in immune evasion (32, 33), and here we have established that residue 197 is important for the binding of an inhibitory, strain-specific MAb, MAb 1F9. Several mutations were created in the context of 3D7 AMA1 domain I; these substitutions in AMA1 were selected to reflect the gene sequences identified in parasite isolates from Africa and Papua New Guinea (8, 9, 32, 33). Since these mutations are found in parasites in the field, the possibility that the overall structure of AMA1 will be disrupted due to the introduction of the mutation is significantly reduced. Of the mutations generated in this study, only those at position 197 resulted in a reduced affinity of MAb 1F9 for AMA1 domain I. With the exception of position 197, no polymorphic residues studied contribute to the MAb 1F9 epitope, as substitution at these positions had little effect on MAb 1F9 binding. One caveat to this was the 242 position, where the replacement of D with Y consistently resulted in an increase in MAb 1F9 binding. Taken together, the experimental information localizes the MAb 1F9 epitope to a short fragment of AMA1 domain I. Five of the six polymorphic residues mutated in this study are located at one end of the fragment, and the sixth residue is located at the other end (Fig. 6A and B). Although residue 197 is a critical component of the epitope, other residues within this fragment must be involved in the overall MAb 1F9 epitope. Further work is under way to identify these residues.
Recently Bai et al. (3) have solved the three-dimensional structure of the domain I+II fragment of P. falciparum AMA1. The 57-residue domain I fragment containing the MAb 1F9 epitope contains the -helix and strands 3 and 4 of the -sheet from the PAN domain of AMA1 domain I identified in the crystal structure (Fig. 6.). Residue 197 is located on a surface-exposed turn of the -helix. The polymorphic residues 196, 242, 243, and 244 that were shown here not to contribute to the 1F9 epitope are in close proximity to residue 197 on the surface of AMA1. However, the structure shows that other polymorphic residues are also in close proximity to residue 197, and it will be a priority to examine the importance of these residues in future studies.
Of the two MAbs used in this study that recognize conformational epitopes on AMA1, one (1F9) was inhibitory whereas the other (2C5) was noninhibitory. The extensively studied 4G2dc1 inhibitory MAb (7, 13, 31) also recognizes a conformationally dependent epitope. Although we have identified residue 197 as a major component of the inhibitory epitope defined by MAb 1F9, until recently a detailed molecular description of the MAb 4G2 epitope has not been available. The broad reactivity of MAb 4G2 with AMA1 molecules from a variety of P. falciparum strains suggests that nonpolymorphic residues contribute substantially to the epitope. Pizarro and colleagues (31) carried out alanine replacements for several residues within P. falciparum AMA1. The residues shown to affect 4G2dc1 binding were identified as nonpolymorphic, and most were localized to a conserved region of domain II in P. falciparum AMA1. It has been previously noted that AMA1 has a "polymorphic" face and a "nonpolymorphic" face, and the positions of the 1F9 and 4G2dcl epitopes on these two faces of the protein are shown in Fig. 6C. Although they are on opposite sides of AMA1, both epitopes flank a deep hydrophobic cleft on the surface of the molecule. It is possible that the interaction of a ligand with this hydrophobic cleft is an important step in the invasion pathway and that binding of either MAb 1F9 or 4G2dc1 could partially obscure this cleft and therefore prevent this interaction. This possible explanation of the invasion-inhibitory properties of these two very different MAbs is currently under investigation in our laboratory.
MAb 1F9 is one of several reagents, including peptides, small proteins, and antibodies, that inhibit invasion and bind to AMA1 (5, 7, 15, 21, 26, 30). Several of these other reagents compete with MAb 1F9 for binding to AMA1 and are therefore likely to bind to the same area of the protein in order to exert their inhibitory effects. Therefore, it is possible that the fragment described in this study is part of a functional "hot spot" on the surface of AMA1, as previously described. It is conceivable that this hot spot is a component of the hydrophobic cleft and may represent an attractive target for small-molecule antimalarial therapy. The identity of residue 197 is clearly important for MAb 1F9 binding, but its importance in a naturally acquired, protective antibody response to AMA1 is unclear. Nevertheless, it would be desirable that an antibody response by an AMA1 vaccine not be dominated by specificities recognizing the MAb 1F9 epitope. With this in mind, consideration should be given to the nature of this and other polymorphic residues in the design and formulation of an AMA1-based vaccine in order to provide protection against a wider cohort of AMA1 genotypes.
The approach taken in this study was to analyze the effect of single polymorphisms introduced into the 3D7 AMA1 domain I on binding of MAb 1F9 when displayed on the surface of M13 bacteriophage. Phage display has been used successfully to identify epitopes in a variety of systems (7, 12, 28), and the technique relies on the faithful display of both continuous and discontinuous epitopes within the proteins displayed on M13 phage. M13 phage display has been shown to allow the formation of even complex epitopes due to the periplasmic folding of proteins during phage production. Previously we have described the display of the complete AMA1 ectodomain in addition to the individual domains on phage and have shown their utility in epitope mapping and structural analyses (7, 29). In this study we have exploited the technology for functional analysis of individual polymorphisms within AMA1. We believe the approach of polymorphism-scanning mutagenesis is a suitable method for determining the effects of individual or multiple polymorphisms on the binding of antibodies to their respective antigens. It obviates the often expensive and time-consuming need to produce multiple recombinant proteins or protein domains, some of which require complex folding and disulfide bonding in order to represent the authentic antigen. Polymorphism scanning also offers a relatively high-throughput procedure for the study of polymorphic residues in vaccine candidate molecules such as P. falciparum AMA1. It is likely that the approach taken here is highly suitable for the analysis of the immunological significance of polymorphisms in other protein antigens, both in malaria parasites and in other pathogens.
ACKNOWLEDGMENTS
This work was supported in part by the National Health and Medical Research Council of Australia and the National Institutes of Health (NIH grant R01AI59229).
REFERENCES
1. Anders, R. F., P. E. Crewther, S. Edwards, M. Margetts, M. L. Matthew, B. Pollock, and D. Pye. 1998. Immunisation with recombinant AMA-1 protects mice against infection with Plasmodium chabaudi. Vaccine 16:240-247.
2. Anders, R. F., D. J. McColl, and R. L. Coppel. 1993. Molecular variation in Plasmodium falciparum: polymorphic antigens of asexual erythrocytic stages. Acta Trop. 53:239-253.
3. Bai, T., M. Becker, A. Gupta, P. Strike, V. J. Murphy, R. F. Anders, and A. H. Batchelor. 2005. Structure of AMA1 from Plasmodium falciparum reveals a clustering of polymorphisms that surround a conserved hydrophobic pocket. Proc. Natl. Acad. Sci. USA 102:12736-12741.
4. Basco, L. K., F. Marquet, M. M. Makler, and J. Le Bras. 1995. Plasmodium falciparum and Plasmodium vivax: lactate dehydrogenase activity and its application for in vitro drug susceptibility assay. Exp. Parasitol. 80:260-271.
5. Casey, J. L., A. M. Coley, R. F. Anders, V. J. Murphy, K. S. Humberstone, A. W. Thomas, and M. Foley. 2004. Antibodies to malaria peptide mimics inhibit Plasmodium falciparum invasion of erythrocytes. Infect. Immun. 72:1126-1134.
6. Chesne-Seck, M. L., J. C. Pizarro, B. V. Normand, C. R. Collins, M. J. Blackman, B. W. Faber, E. J. Remarque, C. H. Kocken, A. W. Thomas, and G. A. Bentley. 2005. Structural comparison of apical membrane antigen 1 orthologues and paralogues in apicomplexan parasites. Mol. Biochem. Parasitol. 144:55-67.
7. Coley, A. M., N. V. Campanale, J. L. Casey, A. N. Hodder, P. E. Crewther, R. F. Anders, L. M. Tilley, and M. Foley. 2001. Rapid and precise epitope mapping of monoclonal antibodies against Plasmodium falciparum AMA1 by combined phage display of fragments and random peptides. Protein Eng. 14:691-698.
8. Cortes, A., M. Mellombo, R. Masciantonio, V. J. Murphy, J. C. Reeder, and R. F. Anders. 2005. Allele specificity of naturally acquired antibody responses against Plasmodium falciparum apical membrane antigen 1. Infect. Immun. 73:422-430.
9. Cortes, A., M. Mellombo, I. Mueller, A. Benet, J. C. Reeder, and R. F. Anders. 2003. Geographical structure of diversity and differences between symptomatic and asymptomatic infections for Plasmodium falciparum vaccine candidate AMA1. Infect. Immun. 71:1416-1426.
10. Crewther, P. E., J. G. Culvenor, A. Silva, J. A. Cooper, and R. F. Anders. 1990. Plasmodium falciparum: two antigens of similar size are located in different compartments of the rhoptry. Exp. Parasitol. 70:193-206.
11. Crewther, P. E., M. L. Matthew, R. H. Flegg, and R. F. Anders. 1996. Protective immune responses to apical membrane antigen 1 of Plasmodium chabaudi involve recognition of strain-specific epitopes. Infect. Immun. 64:3310-3317.
12. Cui, X., H. S. Nagesha, and I. H. Holmes. 2003. Mapping of conformational epitopes on capsid protein VP2 of infectious bursal disease virus by fd-tet phage display. J. Virol. Methods 114:109-112.
13. Dutta, S., J. D. Haynes, A. Barbosa, L. A. Ware, J. D. Snavely, J. K. Moch, A. W. Thomas, and D. E. Lanar. 2005. Mode of action of invasion-inhibitory antibodies directed against apical membrane antigen 1 of Plasmodium falciparum. Infect. Immun. 73:2116-2122.
14. Escalante, A. A., A. A. Lal, and F. J. Ayala. 1998. Genetic polymorphism and natural selection in the malaria parasite Plasmodium falciparum. Genetics 149:189-202.
15. Harris, K. S., J. L. Casey, A. M. Coley, R. Masciantonio, J. K. Sabo, D. W. Keizer, E. F. Lee, A. McMahon, R. S. Norton, R. F. Anders, and M. Foley. 2005. Binding hot spot for invasion inhibitory molecules on Plasmodium falciparum apical membrane antigen 1. Infect. Immun. 73:6981-6989.
16. Healer, J., V. Murphy, A. N. Hodder, R. Masciantonio, A. W. Gemmill, R. F. Anders, A. F. Cowman, and A. Batchelor. 2004. Allelic polymorphisms in apical membrane antigen-1 are responsible for evasion of antibody-mediated inhibition in Plasmodium falciparum. Mol. Microbiol. 52:159-168.
17. Hehl, A. B., C. Lekutis, M. E. Grigg, P. J. Bradley, J. F. Dubremetz, E. Ortega-Barria, and J. C. Boothroyd. 2000. Toxoplasma gondii homologue of plasmodium apical membrane antigen 1 is involved in invasion of host cells. Infect. Immun. 68:7078-7086.
18. Hodder, A. N., P. E. Crewther, and R. F. Anders. 2001. Specificity of the protective antibody response to apical membrane antigen 1. Infect. Immun. 69:3286-3294.
19. Hodder, A. N., P. E. Crewther, M. L. Matthew, G. E. Reid, R. L. Moritz, R. J. Simpson, and R. F. Anders. 1996. The disulfide bond structure of Plasmodium apical membrane antigen-1. J. Biol. Chem. 271:29446-29452.
20. Howell, S. A., F. Hackett, A. M. Jongco, C. Withers-Martinez, K. Kim, V. B. Carruthers, and M. J. Blackman. 2005. Distinct mechanisms govern proteolytic shedding of a key invasion protein in apicomplexan pathogens. Mol. Microbiol. 57:1342-1356.
21. Keizer, D. W., L. A. Miles, F. Li, M. Nair, R. F. Anders, A. M. Coley, M. Foley, and R. S. Norton. 2003. Structures of phage-display peptides that bind to the malarial surface protein, apical membrane antigen 1, and block erythrocyte invasion. Biochemistry 42:9915-9923.
22. Kennedy, M. C., J. Wang, Y. Zhang, A. P. Miles, F. Chitsaz, A. Saul, C. A. Long, L. H. Miller, and A. W. Stowers. 2002. In vitro studies with recombinant Plasmodium falciparum apical membrane antigen 1 (AMA1): production and activity of an AMA1 vaccine and generation of a multiallelic response. Infect. Immun. 70:6948-6960.
23. Kocken, C. H., D. L. Narum, A. Massougbodji, B. Ayivi, M. A. Dubbeld, A. van der Wel, D. J. Conway, A. Sanni, and A. W. Thomas. 2000. Molecular characterisation of Plasmodium reichenowi apical membrane antigen-1 (AMA-1), comparison with P. falciparum AMA-1, and antibody-mediated inhibition of red cell invasion. Mol. Biochem. Parasitol. 109:147-156.
24. Kramer, K. J., S. C. Kan, and W. A. Siddiqui. 1982. Concentration of Plasmodium falciparum-infected erythrocytes by density gradient centrifugation in Percoll. J. Parasitol. 68:336-337.
25. Lambros, C., and J. P. Vanderberg. 1979. Synchronization of Plasmodium falciparum erythrocytic stages in culture. J. Parasitol. 65:418-420.
26. Li, F., A. Dluzewski, A. M. Coley, A. Thomas, L. Tilley, R. F. Anders, and M. Foley. 2002. Phage-displayed peptides bind to the malarial protein apical membrane antigen-1 and inhibit the merozoite invasion of host erythrocytes. J. Biol. Chem. 277:50303-50310.
27. Mascola, J. R., M. K. Louder, T. C. VanCott, C. V. Sapan, J. S. Lambert, L. R. Muenz, B. Bunow, D. L. Birx, and M. L. Robb. 1997. Potent and synergistic neutralization of human immunodeficiency virus (HIV) type 1 primary isolates by hyperimmune anti-HIV immunoglobulin combined with monoclonal antibodies 2F5 and 2G12. J. Virol. 71:7198-7206.
28. Muhle, C., S. Schulz-Drost, A. V. Khrenov, E. L. Saenko, J. Klinge, and H. Schneider. 2004. Epitope mapping of polyclonal clotting factor VIII-inhibitory antibodies using phage display. Thromb. Haemost. 91:619-625.
29. Nair, M., M. G. Hinds, A. M. Coley, A. N. Hodder, M. Foley, R. F. Anders, and R. S. Norton. 2002. Structure of domain III of the blood-stage malaria vaccine candidate, Plasmodium falciparum apical membrane antigen 1 (AMA1). J. Mol. Biol. 322:741-753.
30. Nuttall, S. D., K. S. Humberstone, U. V. Krishnan, J. A. Carmichael, L. Doughty, M. Hattarki, A. M. Coley, J. L. Casey, R. F. Anders, M. Foley, R. A. Irving, and P. J. Hudson. 2004. Selection and affinity maturation of IgNAR variable domains targeting Plasmodium falciparum AMA1. Proteins 55:187-197.
31. Pizarro, J. C., B. Vulliez-Le Normand, M. L. Chesne-Seck, C. R. Collins, C. Withers-Martinez, F. Hackett, M. J. Blackman, B. W. Faber, E. J. Remarque, C. H. Kocken, A. W. Thomas, and G. A. Bentley. 2005. Crystal structure of the malaria vaccine candidate apical membrane antigen 1. Science 308:408-411.
32. Polley, S. D., W. Chokejindachai, and D. J. Conway. 2003. Allele frequency-based analyses robustly map sequence sites under balancing selection in a malaria vaccine candidate antigen. Genetics 165:555-561.
33. Polley, S. D., and D. J. Conway. 2001. Strong diversifying selection on domains of the Plasmodium falciparum apical membrane antigen 1 gene. Genetics 158:1505-1512.
34. Polley, S. D., T. Mwangi, C. H. Kocken, A. W. Thomas, S. Dutta, D. E. Lanar, E. Remarque, A. Ross, T. N. Williams, G. Mwambingu, B. Lowe, D. J. Conway, and K. Marsh. 2004. Human antibodies to recombinant protein constructs of Plasmodium falciparum apical membrane antigen 1 (AMA1) and their associations with protection from malaria. Vaccine 23:718-728.
35. Sidhu, S. S., H. B. Lowman, B. C. Cunningham, and J. A. Wells. 2000. Phage display for selection of novel binding peptides. Methods Enzymol. 328:333-363.
36. Silvie, O., J. F. Franetich, S. Charrin, M. S. Mueller, A. Siau, M. Bodescot, E. Rubinstein, L. Hannoun, Y. Charoenvit, C. H. Kocken, A. W. Thomas, G. J. Van Gemert, R. W. Sauerwein, M. J. Blackman, R. F. Anders, G. Pluschke, and D. Mazier. 2004. A role for apical membrane antigen 1 during invasion of hepatocytes by Plasmodium falciparum sporozoites. J. Biol. Chem. 279:9490-9496.
37. Snow, R. W., C. A. Guerra, A. M. Noor, H. Y. Myint, and S. I. Hay. 2005. The global distribution of clinical episodes of Plasmodium falciparum malaria. Nature 434:214-217.
38. Stowers, A. W., M. C. Kennedy, B. P. Keegan, A. Saul, C. A. Long, and L. H. Miller. 2002. Vaccination of monkeys with recombinant Plasmodium falciparum apical membrane antigen 1 confers protection against blood-stage malaria. Infect. Immun. 70:6961-6967.
39. Trager, W., and J. B. Jensen. 1976. Human malaria parasites in continuous culture. Science 193:673-675.(A. M. Coley, K. Parisi, R)