Whole-genome surveillance identifies markers of Plasmodium falciparum drug resistance and novel genomic regions under selection in Mozambique

Abstract

Malaria is a devastating disease caused by Plasmodium parasites. The evolution of parasite drug resistance continues to hamper progress toward malaria elimination, and despite extensive efforts to control malaria, it remains a leading cause of death in Mozambique and other countries in the region. The development of successful vaccines and identification of molecular markers to track drug efficacy are essential for managing the disease burden. We present an analysis of the parasite genome in Mozambique, a country with one of the highest malaria burdens globally and limited available genomic data, revealing current selection pressure. We contribute additional evidence to limited prior studies supporting the effectiveness of SWGA in producing reliable genomic data from complex clinical samples. Our results provide the identity of genomic loci that may be associated with current antimalarial drug use, including artemisinin and lumefantrine, and reveal selection pressure predicted to compromise the efficacy of current vaccine candidates.

Keywords: Plasmodium falciparum; antimalarial drug resistance; genetic variation; positive selection; selective whole genome amplification; severe malaria; whole-genome sequencing.

Fig 1

Validation of Mozambique data set. (a) Types of genic and intergenic variants detected as a homozygous call in at least one Mozambique sample (136,720), or detected at least five times (high probability, 27,070 variants) in one of the 120 samples. (b) Proportions of missense and synonymous (syn.) mutations in all (61,392) and high probability coding variants (10,567) in Mozambique samples and in variants identified in PlasmoDB release 61 (333,648 and 239,283, respectively) for 5,361 genes (mitochondrial, RNA, apicoplast, and pseudogenes removed). Genes were classified based on essentiality (25) as well as the presence of recognizable PFAM domains and were normalized based on total genes in the category. The data show a higher proportion of mutations in nonessential genes with no homology to other genes, typically classified as “conserved, unknown function” and a lower proportion in genes such as enzymes. Noncore genes were defined by Otto et al. (26) and consist largely of var genes, rifins and stevors (281 genes). (c) Principal component analysis (PCA) of Mozambique variants in relationship to variants from the Pf6 data set, comprising genome sequences for 7,113 P. falciparum samples from 28 malaria-endemic countries spanning 2001–2015 (7). PCA was performed on binary encoding of alternate allele presence in 459,725 core genome sites that were called as potential missense variants in Pf6 and were classified as present in at least one sample from the combined Mozambique and Pf6 data (see Materials and Methods).

Fig 2

Selection signals detected via identity by descent and extended haplotype homozygosity. (a) The isoRelate R package was used to scan for regions of the genome with a high proportion of sample pairs that were IBD and to calculate the statistical significance of the IBD sharing. (b) The upper x-axes are annotated with the locations of genes previously found to be under selection. Significant IBD sharing (−log₁₀ P > 5) was detected near pfaat1 and regions of chromosomes 13 and 14 (candidate genes listed in Table 3). (c) The extended haplotype homozygosity in pfaat1 decays more slowly around the S258L allele (red) relative to the ancestral allele (blue), suggesting that it is under selection.

Fig 3

Mutations in pfaat1 and pfmrp2 observed in field studies and in vitro evolution experiments. (a) AlphaFold homology model of pfaat1. Mutations from in vitro evolution experiments are highlighted in purple (59). Mutations observed in field isolates from the Pf6 database are highlighted in orange (58). Mutations observed in Mozambique are highlighted in red. The S258L mutation was present in both Mozambique and field isolates from the Pf6 database. (b) AlphaFold homology model of pfmrp2 highlighted with mutations observed in Mozambique field isolates. (c and d) TOPCONS topology models show that many of the observed mutations are in the transmembrane channel portions of the pfaat1 and pfmrp2 proteins, respectively.

Fig 4

Genes under selection as determined by Tajima’s D and dN/dS scores. (a) Tajima’s D values plotted by chromosome to reveal genes under balancing selection, revealing high scores for targets of vaccines in development and testing. Drug resistance genes studied have moderately negative scores as expected since high mutation rates in single or only a few SNPs are not expected to drive large Tajima’s D scores. They are reported here: pfaat1 (0.102979664), pfdhfr (−1.1736371), pfdhps (−1.040123), pfcrt (−1.5699201), pfmdr1 (−0.20613), and pfkelch13 (−1.3879097). (b) Missense mutations identified in csp map to T-cell epitopes. (c) dN/dS (ratio of missense [nonsynonymous{NS}] to synonymous SNV counts normalized by expected numbers of NS and S sites under a neutral model) for 94 Mozambique samples with sufficient coverage vs 5,969 QC pass Pf6 samples (excluding samples from Mozambique) for all genes with at least one SNV (4,468 genes). Ratios are estimated using pseudocounts for genes with no synonymous changes.