Molecular epidemiology of malaria

Abstract

Malaria persists as an undiminished global problem, but the resources available to address it have increased. Many tools for understanding its biology and epidemiology are well developed, with a particular richness of comparative genome sequences. Targeted genetic manipulation is now effectively combined with in vitro culture assays on the most important human parasite, Plasmodium falciparum, and with in vivo analysis of rodent and monkey malaria parasites in their laboratory hosts. Studies of the epidemiology, prevention, and treatment of human malaria have already been influenced by the availability of molecular methods, and analyses of parasite polymorphisms have long had useful and highly informative applications. However, the molecular epidemiology of malaria is currently undergoing its most substantial revolution as a result of the genomic information and technologies that are available in well-resourced centers. It is a challenge for research agendas to face the real needs presented by a disease that largely exists in extremely resource-poor settings, but it is one that there appears to be an increased willingness to undertake. To this end, developments in the molecular epidemiology of malaria are reviewed here, emphasizing aspects that may be current and future priorities.

FIG. 1.

Analysis of the nonrepeat region of the csp gene of P. knowlesi parasites from 8 human infections (representative of 106 human P. knowlesi infections detected) in Kapit District Hospital (KH), Sarawak, Malaysia. (A) Phylogenetic analysis of the human isolates together with 2 P. knowlesi macaque isolates and other malaria parasite species. The large clade at the top contains P. vivax and the Asian macaque parasites, with P. knowlesi human and macaque sequences clearly distinct from all other species (numbers on the branches show bootstrap support, which is 100% in this case). (B) Sequence differences among the P. knowlesi isolates indicates a normal level of within-species polymorphism in the csp gene and strongly indicates that the human isolates have not arisen from a single recent source. Similar results are also obtained by studying the small subunit rRNA gene sequence in the same samples. (Reprinted from reference with permission from Elsevier.)

FIG. 2.

Schematic diagram indicating fluctuation in peripheral blood parasitemia of P. falciparum relative to an arbitrary threshold of detectability (10⁶ parasites in the body). Three coinfecting genotypes (A to C) are considered. All genotypes show typical 2-day periodicity of detectable levels due to sequestration of mature asexual developmental stages that disappear from the peripheral blood. Genotype A declines over time, genotype B remains at a constant level throughout, while genotype C declines initially and then increases over time. The arrows show the genotypes that would be detected in peripheral blood samples taken on days 1, 6, 11, and 16, with all samples giving a different result. An incorrect interpretation would be that genotypes B and C appeared later as “new infections.”

FIG. 3.

P. falciparum parasite populations sampled in different areas of Malaysian Borneo show divergence in allele frequencies consistent with an isolation by distance model. (A) Map of 8 locations from which 288 P. falciparum infections were sampled for genotyping of 10 microsatellite loci. (B) Pairwise genetic distances between populations estimated by the transformed fixation index F_ST/(1 − F_ST) correlates with geographical distance (natural log of distance in km). (Reprinted from reference with permission. © 2005 by the Infectious Diseases Society of America. All rights reserved.)

FIG. 4.

Microsatellite variability is reduced around the dhfr locus on P. falciparum chromosome 4 in a population sample from the Thai-Myanmar border. Diversity is plotted on the y axis as the expected heterozygosity index (with 1 standard error). Distances (in kb) of microsatellites from the dhfr gene are shown on the x axis. The observed reduction in diversity is due to a selective sweep that caused a drug resistance allele encoded by the dhfr gene to become very common in the population and fits expectations under population genetic models with realistic parameters, as shown by the black and dotted lines. (Reprinted from reference by permission of Oxford University Press.)