Evaluation of the antigenic diversity of placenta-binding Plasmodium falciparum variants and the antibody repertoire among pregnant women

Abstract

Pregnant women are infected by specific variants of Plasmodium falciparum that adhere and accumulate in the placenta. Using serological and molecular approaches, we assessed the global antigenic diversity of surface antigens expressed by placenta-binding isolates to better understand immunity to malaria in pregnancy and evolution of polymorphisms and to inform vaccine development. We found that placenta-binding isolates originating from all major regions where malaria occurs were commonly recognized by antibodies in different populations of pregnant women. There was substantial antigenic overlap and sharing of epitopes between isolates, including isolates from distant geographic locations, suggesting that there are limitations to antigenic diversity; however, differences between populations and isolates were also seen. Many women had cross-reactive antibodies and/or a broad repertoire of antibodies to different isolates. Studying VAR2CSA as the major antigen expressed by placenta-binding isolates, we identified antibody epitopes encoded by variable sequence blocks in the DBL3 domain. Analysis of global var2csa DBL3 sequences demonstrated that there was extensive sharing of variable blocks between Africa, Asia, Papua New Guinea, and Latin America, which likely contributes to the high level of antigenic overlap between different isolates. However, there was also evidence of geographic clustering of sequences and differences in VAR2CSA sequences between populations. The results indicate that there is limited antigenic diversity in placenta-binding isolates and may explain why immunity to malaria in pregnancy can be achieved after exposure during one pregnancy. Inclusion of a limited number of variants in a candidate vaccine may be sufficient for broad population coverage, but geographic considerations may also have to be included in vaccine design.

FIG. 1.

var2csa transcription by various CSA-binding parasite isolates. (A) Origins of isolates used in this study. (B and C) RNA from ring-stage parasites was detected with a var2csa-specific probe (B) or var exon 2, which binds to most var genes (C). The dominant transcript in all isolates is var2csa. Note the presence of the double band with the var2csa-specific probe, indicating expression of two variants of var2csa. In both blots Pf2004-CSA was exposed slightly longer due to its lower RNA content. In addition to the parental HCS3 isolate, two clones of HCS3 parasites, E5 and G7, were included.

FIG. 2.

Antibodies to a global selection of placenta-type CSA-binding isolates in malaria-exposed pregnant women in Malawi and PNG. (A) Parasite isolates from different parts of the world selected for CSA adhesion were tested with sera from malaria-exposed women in Malawi by using FACS. The results of one of two independent experiments for each isolate are shown. The geometric mean fluorescence intensities (MFI) for IgG to VSA in sera from aparasitemic (apara) and parasitemic (para) PG and MG women, men, and Melbourne controls are shown. Isolates were recognized in a gender- and gravidity-dependent fashion. Many samples from pregnant women had antibodies to the isolates, whereas there was no or little reactivity for men (with the exception of Pf2006-CSA) and there was no reactivity for Melbourne controls. Aparasitemic MG women showed higher median MFI (horizontal lines) than aparasitamic PG women and men. (B) A global panel of P. falciparum isolates was selected for CSA adhesion and tested with sera from pregnant women and men from PNG and Melbourne controls by using FACS. The results of one of two independent experiments for each isolate are shown. Isolates were recognized in a gender- and gravidity-dependent fashion. For each isolate, a subset of pregnant women had antibodies, whereas the sera of men exhibited little or no recognition and the sera of Melbourne controls uniformly lacked antibodies. The horizontal lines indicate the median MFI. Note that in some cases the y axis starts below zero so that all data points can be shown.

FIG. 3.

Antigenic overlap and differences between isolates of different geographic origins. The correlations between antibody reactivities to different isolates are shown for selected isolate combinations. Antibodies to isolates were measured by FACS for samples from Malawi (A, C, and E) and PNG (B, D, and F). Data points represent the mean values resulting from testing samples with each isolate in two independent experiments. All correlations were calculated by using Spearman's rho and are statistically significant (P < 0.0001), except for the correlations shown in panels C (P = 0.142) and E (P = 0.092).

FIG. 4.

Cross-reactive and variant-specific antibodies in pregnant women detected using mixed-agglutination assays. Serum samples from Malawian women were tested for the ability to agglutinate isolates in mixed-agglutination assays. (A) Percentages of agglutinates that were mixed agglutinates for a selection of isolate combinations. The values are the means resulting from testing serum samples from different women. Mixed agglutinates were defined as agglutinates containing ≥2 IEs of each parasite line and indicate the presence of cross-reactive antibodies to the two isolates. (B) Typical mixed-color agglutinate of two isolates, resulting from the presence of cross-reactive antibodies. The two isolates were CS2 (DAPI, blue) and HCS3 (ethidium bromide, red). (C) Samples from primigravid (PG), secundigravid (SG), and multigravid (MG) women from Malawi were tested for the ability to agglutinate CS2 and HCS3. Shown are the proportions of agglutinates that were mixed agglutinates (the horizontal lines indicate the medians).

FIG. 5.

Polymorphic loops formed by variable sequence blocks of DBL3 were recognized by vaccine-induced and acquired antibodies. (A) Rabbit antiserum raised against the DBL1 or DBL3 domain of IT4 VAR2CSA was tested by ELISA for reactivity against synthetic peptides, which represent polymorphic loops of the DBL3 domain (IT4). Antibodies generated by DNA vaccination (aDBL3 DNA) recognized only loop 5, whereas antibodies generated by immunization with recombinant protein (aDBL3 Pp) reacted with IT4 DBL3 loops 5 and 3. Negative control sera against the VAR2CSA DBL1 domain were generated by DNA vaccination (aDBL1 DNA) or with recombinant protein (aDBL1 Pp) and did not react with either DBL3 loop. Subsequently, human sera from malaria-exposed Malawian (n = 72) and PNG (n = 56) women were tested for antibodies to IT4 DBL3 loop 5 peptide by an ELISA. The peptide was recognized by PG women, MG women, and several sympatric men from both countries, but not by Melbourne controls. Results for a representative selection of sera from Malawi (B) are shown. All women included were in the top quartile of responders to CS2 in FACS-based assays (see Table 1). In panel B samples 27, 71, 98, and 161 were samples from MG women, samples 66 and 145 were samples from PG women, and sample c10 was a sample from a Melbourne control donor. All samples were run in duplicate; means and standard deviations are shown. FB, final bleed; PI, preimmunization serum; OD (405), optical density at 405 nm. (C and D) Identical or nearly identical variable block 5 (C) and variable block 3 (D) sequences could be identified for isolates from diverse geographic origins. Each group contains sequences from isolates from different geographic regions, and the sequences were grouped using sequences obtained from one of the placenta-binding isolates used in antibody studies (3D7-CSA, CS2/IT4, HCS3, XIE-CSA, HB3-CSA, Pf2004-CSA, and Pf2006-CSA). Sequence alignments were constructed on the basis of identity in the region corresponding to synthetic peptides used in immunoassays (A and B). Blue shading indicates the predicted loop (additional residues are also shown), and red shading indicates an amino acid change compared to the majority of the sequences. The overall sequence alignment is indicated by bars; red indicates a 100% match, while orange, green, and blue indicate increasing numbers of sequence differences. The origins of the parasites are as follows: CYK34.1, CYK37, and CYK42.1, Senegal; D10, PNG; MC, Malaysia; Mplc21-1, Malawi; PC49, Peru; T2C6, Thailand; XHA clones, PNG; XIE par (non-CSA-selected XIE clones), PNG; XJA clones, PNG.

FIG. 6.

Globally diverse isolates share DBL3 variable block sequences. Analysis of VAR2CSA DBL3 variable block 3 and 5 amino acid sequences demonstrated that isolates having diverse geographic origins shared the same polymorphic loop sequences. A diagram of the DBL3 domain is shown at the top. The positions and lengths of variable blocks 1 to 5 in the DBL3 sequences are indicated. The locations of primers used for PCR amplification are indicated by arrowheads. The filled arrowheads indicate primers used to amplify PNG sequences in this study, and the open arrowheads indicate the locations of primers used previously to amplify sequences from Senegal isolates (1). The amino acid position of the primers is counted from the first amino acid in the IT4 VAR2CSA sequence (accession no. AAQ73926). The segmentation structures for variable blocks 3 and 5 in representative PNG sequences are shown below the diagram of the DBL3 domain (a full alignment of all sequences is shown in Fig. S1 in the supplemental material). Variable block 3 is composed of three segments, segments a, b, and c, and variable block 5 is composed of two segments, segments a and b. The locations of variable block peptides used for serological analysis are indicated under the alignments. PNG sequences that were identical to the IT4 VAR2CSA peptide sequences are indicated by filled circles.

FIG. 7.

Overlap and differences in distribution of polymorphic segments in variable blocks 3 and 5. The major segment types found in variable blocks 3 and 5 in the VAR2CSA DBL3 domain were quantified. A total of 124 DBL3 sequences were compared (16 sequences from Southeast Asia, 54 sequences from PNG, 43 sequences from Senegal, 8 sequences from other African countries, and 3 sequences from Central and South America [C/S America], as well as one sequence of unknown origin (3D7). (A) Major segment types of vb3 segment a. (B) Major segment types of vb3 segment b. (C) Major segment types of vb3 segment c. (D) Major segment types of vb5 segment a. (E) Major segment types of vb5 segment b. When data for PNG (n = 54) and Senegal (n = 43) were compared, there were statistically significant differences in the distribution of sequences in variable block 3 and 5 (P < 0.0001 for differences in the proportions of vb3 segment a and vb3 segment b sequences; P = 0.045 for differences in the proportions of vb3 segment c sequences; P = 0.061 and P = 0.023 for differences in the proportions of vb5 segment a and vb5 segment b sequences, respectively [Fisher's exact test]).