On the Evolution and Function of Plasmodium vivax Reticulocyte Binding Surface Antigen (pvrbsa)

Abstract

The RBSA protein is encoded by a gene described in Plasmodium species having tropism for reticulocytes. Since this protein is antigenic in natural infections and can bind to target cells, it has been proposed as a potential candidate for an anti-Plasmodium vivax vaccine. However, genetic diversity (a challenge which must be overcome for ensuring fully effective vaccine design) has not been described at this locus. Likewise, the minimum regions mediating specific parasite-host interaction have not been determined. This is why the rbsa gene's evolutionary history is being here described, as well as the P. vivax rbsa (pvrbsa) genetic diversity and the specific regions mediating parasite adhesion to reticulocytes. Unlike what has previously been reported, rbsa was also present in several parasite species belonging to the monkey-malaria clade; paralogs were also found in Plasmodium parasites invading reticulocytes. The pvrbsa locus had less diversity than other merozoite surface proteins where natural selection and recombination were the main evolutionary forces involved in causing the observed polymorphism. The N-terminal end (PvRBSA-A) was conserved and under functional constraint; consequently, it was expressed as recombinant protein for binding assays. This protein fragment bound to reticulocytes whilst the C-terminus, included in recombinant PvRBSA-B (which was not under functional constraint), did not. Interestingly, two PvRBSA-A-derived peptides were able to inhibit protein binding to reticulocytes. Specific conserved and functionally important peptides within PvRBSA-A could thus be considered when designing a fully-effective vaccine against P. vivax.

Keywords: Plasmodium vivax; antimalarial vaccine; evolutionary forces; genetic diversity; parasite–host interaction; protein biding; rbsa.

FIGURE 1

Phylogenetic tree mapping the presence/absence of rbsa locus and duplication events. The phylogenetic tree represents Plasmodium evolutionary history. The branches are colored regarding their vertebrate host. The rbsa locus was found in parasites from monkey-malaria clade, but it was not clear whether the rbsa gene became lost from the other clades or arose exclusively in the monkey-malaria clade. rbsa became lost in P. knowlesi and P. coatneyi whilst duplication events occurred in P. vivax and P. cynomolgi. Models of putative ancestral chromosome 3 are displayed, indicating the arrangement after the duplication event (chr derived). PlasmoDB genome fragment annotation is also shown. P. cynomolgi rbsa (pcrbsa) and its paralog (pcrbsap) had a premature stop codon regarding P. vivax rbsa sequences, indicated by a white rectangle in the pcrbsa and pcrbsap gene models.

FIGURE 2

Identifying natural selection signatures in the rbsa gene. (A) A phylogeny tree inferred from pvrbsa CDS and its orthologous CDS was analyzed by the aBSREL method. The shade of each color on branches indicates selection strength [red indicates positive selection (ω > 1) and green negative selection (ω < 1)]. The size of each colored segment represents the percentage of selected sites in the corresponding ω class. The Figure shows the ω values, the percentage of positive selected sites (Pr [ω = ω+]) and p-values. Branches have been classified as undergoing episodic diversifying selection, by the p-value corrected for multiple testing, using the Holm–Bonferroni method at p < 0.05. (B) Representation of the rbsa encoding DNA sequence and the sliding window for the ω rate. The ω rates within P. vivax (d_N/d_S) and between the species (K_N/K_S) are shown. A gene model of the rbsa gene is given below the sliding window, indicating the sites under purifying selection (green) and positive selection (red). Gene model and numbering were based on the Supplementary Data Sheet 1E alignment.

FIGURE 3

Median-joining network for Colombian and Venezuelan subpopulations. The Figure shows the pvrbsa haplotypes identified from Colombian and Venezuelan isolates. Some haplotypes were included within another haplotype using the contraction star algorithm (Forster et al., 2001) for simplifying network interpretation. Each node is a haplotype and its size indicates its frequency. The lines connecting the haplotypes represent the different mutational paths and the median vectors are the ancestral sequences explaining the relationship and evolutionary origin.

FIGURE 4

PvRBSA reticulocyte binding activity. (A) Purification of full rPvRBSA (F) or its fragments (A and B). The Figure shows each purified molecule analysed by Coomassie blue staining (lines 2, 4, and 6) or Western blot (lines 3, 5, and 7). MW, molecular weight marker (line 1). (B) rPvRBSA target cell binding assay. DBP-RII and -RIII/IV indicates positive and negative binding controls, respectively. (C) Binding inhibition assay. The Figure shows the rPvRBSA reticulocyte binding inhibition percentage using different PvRBSA-A-derived peptides as well as a M. tuberculosis peptide (negative controls). Standard deviation is shown for each assay. C+, shows the assay just with the rPvRBSA protein (positive control). P1-P6 indicate peptides 40893, 40894, 40895, 40896, 40897, and 40898, respectively. P7 is a M. tuberculosis peptide (39266) used as negative control. All peptides were pre-incubated with cells before incubating with the complete protein.