Genomic insights into Plasmodium vivax population structure and diversity in central Africa

Abstract

Background: Though Plasmodium vivax is the second most common malaria species to infect humans, it has not traditionally been considered a major human health concern in central Africa given the high prevalence of the human Duffy-negative phenotype that is believed to prevent infection. Increasing reports of asymptomatic and symptomatic infections in Duffy-negative individuals throughout Africa raise the possibility that P. vivax is evolving to evade host resistance, but there are few parasite samples with genomic data available from this part of the world.

Methods: Whole genome sequencing of one new P. vivax isolate from the Democratic Republic of the Congo (DRC) was performed and used in population genomics analyses to assess how this central African isolate fits into the global context of this species.

Results: Plasmodium vivax from DRC is similar to other African populations and is not closely related to the non-human primate parasite P. vivax-like. Evidence is found for a duplication of the gene PvDBP and a single copy of PvDBP2.

Conclusion: These results suggest an endemic P. vivax population is present in central Africa. Intentional sampling of P. vivax across Africa would further contextualize this sample within African P. vivax diversity and shed light on the mechanisms of infection in Duffy negative individuals. These results are limited by the uncertainty of how representative this single sample is of the larger population of P. vivax in central Africa.

Keywords: Central Africa; Duffy negative; Genome; Malaria; Plasmodium vivax; Sub-Saharan Africa.

Fig. 1

DRC P. vivax falls within African variation in global species context. DRC sample indicated by arrowhead in each panel. A Principal components analysis of global genetic diversity reveals P. vivax from the DRC grouped with African and South Asian populations. The first principal component explains 14.3% of variation in the data and the second principal component represents 10.3% of variation. B Maximum Likelihood tree shows P. vivax from the DRC clusters with Uganda and Madagascar. Tree constructed using whole-genome SNP data and rooted using two P. vivax-like sequences (red dotted line). Continent colours: Americas (green), Africa (blue), South Asia (pink), East Asia (red), Southeast Asia (brown). Tree constructed using IQtree with the GTR + ASC model to account for ascertainment bias in SNP data. Population-level SH-aLRT and UFBoot support values generated by IQTree are shown on the node in the format: SH-aLRT support (%)/ultrafast bootstrap support (%). C Admixture analysis of global population structure indicates P. vivax ancestry is geographically structured. Each vertical bar represents the proportion of genetic ancestry belonging to one individual P. vivax sample for each simulated population size (K). Population size K = 14 is most-supported by Cross Validation Error. D F4 statistics calculated using Admixtools2 using the relationship (P. vivax-like, DRC; Papua New Guinea (PNG), test). Higher F4 estimates indicate DRC has a closer relationship with the test population than it does to the PNG population. Error bars indicate ± 3 SE

Fig. 2

Maximum likelihood tree shows DRC P. vivax sample branches with P. vivax populations and not P. vivax-like. Maximum Likelihood tree of nuclear genome SNP data shows that P. vivax from the DRC does not branch with P. vivax-like samples (black tip labels). The DRC sample, indicated here with an arrow, clusters with other African samples

Fig. 3

Read Pileup image of duplication of PvDBP in P. vivax from DRC. IGV view of the read depth for PvDBP in the DRC P. vivax sample where genome coordinate is on the X-axis and read depth is on the Y-axis. This shows an increase in read depth around PvDBP indicating a duplication, and the different variants present along this region suggest two distinct haplotypes