Contrasting population structures of the genes encoding ten leading vaccine-candidate antigens of the human malaria parasite, Plasmodium falciparum

Abstract

The extensive diversity of Plasmodium falciparum antigens is a major obstacle to a broadly effective malaria vaccine but population genetics has rarely been used to guide vaccine design. We have completed a meta-population genetic analysis of the genes encoding ten leading P. falciparum vaccine antigens, including the pre-erythrocytic antigens csp, trap, lsa1 and glurp; the merozoite antigens eba175, ama1, msp's 1, 3 and 4, and the gametocyte antigen pfs48/45. A total of 4553 antigen sequences were assembled from published data and we estimated the range and distribution of diversity worldwide using traditional population genetics, Bayesian clustering and network analysis. Although a large number of distinct haplotypes were identified for each antigen, they were organized into a limited number of discrete subgroups. While the non-merozoite antigens showed geographically variable levels of diversity and geographic restriction of specific subgroups, the merozoite antigens had high levels of diversity globally, and a worldwide distribution of each subgroup. This shows that the diversity of the non-merozoite antigens is organized by physical or other location-specific barriers to gene flow and that of merozoite antigens by features intrinsic to all populations, one important possibility being the immune response of the human host. We also show that current malaria vaccine formulations are based upon low prevalence haplotypes from a single subgroup and thus may represent only a small proportion of the global parasite population. This study demonstrates significant contrasts in the population structure of P. falciparum vaccine candidates that are consistent with the merozoite antigens being under stronger balancing selection than non-merozoite antigens and suggesting that unique approaches to vaccine design will be required. The results of this study also provide a realistic framework for the diversity of these antigens to be incorporated into the design of next-generation malaria vaccines.

Figure 1. The relationship between polymorphism and haplotype diversity of the genes encoding ten P. falciparum vaccine antigens.

Non-merozoite antigens are represented by a solid symbol and merozoite antigens by an open symbol.

Figure 2. Global population structure of the genes encoding ten P. falciparum vaccine antigens based on Bayesian cluster analysis.

Membership coefficients for A–J) individual nsSNP haplotypes and K–T) the population average for the estimated number of clusters (K, shown on the left of the two histograms). In the latter, countries from different continents are separated by a blank space and organised from east on the left, to west on the right with vaccine haplotypes on the far right hand side. An asterisk denotes countries for which fewer than 8 haplotypes were available that were taken from dataset 2 (Table S2). Dark blue = cluster 1; Red = cluster 2; Green = cluster 3; Purple = cluster 4; Light blue = cluster 5; Orange = cluster 6.

Figure 3. Global population structure of the genes encoding ten P. falciparum vaccine antigens based on network analysis.

Networks of nsSNP haplotypes were drawn by first removing multiple copies of each haplotype , leaving only one copy per country for the analysis. Hence, identical haplotypes from different regions, but not within regions were included. Each node (coloured circle) represents a haplotype, shaded by region of origin: Red = Africa, Green = Asia, Blue = Americas. Nodes are tied by edges (black lines) demonstrating that they share a predefined threshold (t) of nsSNPs for csp = 24; trap = 67; lsa1 = 8; ama1 = 48; eba175 = 18; msp1 = 5; msp3 = 63; msp4 = 19; glurp = 18; pfs48/45 = 23. Vaccine haplotypes are shaded in yellow. Haplotypes originating from isolates with unknown origin are shaded in white (unless they were vaccine haplotypes).