A Worldwide Map of Plasmodium falciparum K13-Propeller Polymorphisms

Abstract

Background: Recent gains in reducing the global burden of malaria are threatened by the emergence of Plasmodium falciparum resistance to artemisinins. The discovery that mutations in portions of a P. falciparum gene encoding kelch (K13)-propeller domains are the major determinant of resistance has provided opportunities for monitoring such resistance on a global scale.

Methods: We analyzed the K13-propeller sequence polymorphism in 14,037 samples collected in 59 countries in which malaria is endemic. Most of the samples (84.5%) were obtained from patients who were treated at sentinel sites used for nationwide surveillance of antimalarial resistance. We evaluated the emergence and dissemination of mutations by haplotyping neighboring loci.

Results: We identified 108 nonsynonymous K13 mutations, which showed marked geographic disparity in their frequency and distribution. In Asia, 36.5% of the K13 mutations were distributed within two areas--one in Cambodia, Vietnam, and Laos and the other in western Thailand, Myanmar, and China--with no overlap. In Africa, we observed a broad array of rare nonsynonymous mutations that were not associated with delayed parasite clearance. The gene-edited Dd2 transgenic line with the A578S mutation, which expresses the most frequently observed African allele, was found to be susceptible to artemisinin in vitro on a ring-stage survival assay.

Conclusions: No evidence of artemisinin resistance was found outside Southeast Asia and China, where resistance-associated K13 mutations were confined. The common African A578S allele was not associated with clinical or in vitro resistance to artemisinin, and many African mutations appear to be neutral. (Funded by Institut Pasteur Paris and others.).

Figure 1. K13 Nonsynonymous Mutations, According to Country and Continent

Shown are the percentages of nonsynonymous mutations that have been identified in the portion of the Plasmodium falciparum K13 gene encoding the kelchpropeller domain in Asia, Africa, South America, and Oceania. Synonymous mutations that do not modify the protein sequence are not indicated. At present, all the mutations that have been associated with resistance to artemisinin derivatives have resulted in nonsynonymous amino acid changes. The black circles indicate medians and the I bars interquartile ranges for each continent. K13 nonsynonymous mutations were not detected in 27 countries from which samples were obtained (19 in Africa, 2 in Asia, 1 in Oceania, and 5 in South America). Names are not shown (owing to a lack of space) for the following countries: in Asia: Afghanistan, Iran, Bangladesh, Nepal, Indonesia, and Philippines; in Africa: Cameroon, Congo, Democratic Republic of Congo, Equatorial Guinea, Gabon, Burundi, Ethiopia, Rwanda, Sudan, South Sudan, Somalia, Tanzania, Uganda, Madagascar, Angola, Malawi, Mozambique, South Africa, Zimbabwe, Benin, Burkina Faso, Ghana, Guinea Bissau, Ivory Coast, Liberia, Mali, Mauritania, Niger, Nigeria, Senegal, Sierra Leone, and Togo; in South America: Brazil, Colombia, Ecuador, French Guiana, Peru, and Venezuela; and in Oceania: Papua New Guinea and Solomon Islands. The percentages for Chad and Gambia were derived from a sample size of less than 50. Details regarding sampling and K13 diversity according to country are provided in Tables S4, S5, and S7 and Fig. S3 in the Supplementary Appendix. CAR denotes Central African Republic.

Figure 2. Frequency Distribution of the Wild-Type K13 Allele

Shown are the distributions of the wild-type K13 allele in Asia (Panel A) and around the world (Panel B). Areas in which malaria is endemic are shaded in gray, and areas that are considered to be malaria-free are shown in white. The mean frequency of the wild-type allele is indicated by the color code. In Panel A, the individual sites of sample collection are indicated with a cross. In Panel B, a 100-km radius was used for the area centered on each sampling site or on the capital city of the country if no specific site was used for sampling. Data regarding sampling methods and K13 diversity according to country are provided in Tables S4, S5, and S7 and Fig. S3 in the Supplementary Appendix.

Figure 3. Overview of the Distribution of the Flanking Haplotypes of the C580Y, Y493H, R539T, I543T, P553L, P574L, and F446I Nonsynonymous Mutations in K13 in Two Regions in Asia

Shown are two areas in Asia in which individual K13 nonsynonymous mutations (which are labeled in the colored boxes) were frequently identified: one area that includes Cambodia, Vietnam, and Laos and a second area that includes western Thailand, Myanmar, and China. The haplotype groups of C580Y, F446I, P574L, and P553L are shown in distinct colors. Similarly colored boxes indicate shared haplotypes, and hatched boxes indicate the presence of additional countryspecific haplotypes. For example, C580Y haplotypes that are shared between Cambodia, Vietnam, and Laos (H29, H33, or H43) are shown in red, whereas the C580Y haplotype found in Thailand (H42) is shown in light blue. Full details about the distribution of haplotypes are provided in Table S9 and Fig. S9 in the Supplementary Appendix.

Figure 4. Algorithm for Surveillance of Artemisinin Resistance on the Basis of K13 Mutations

Shown is a step-by-step procedure for acquiring information to help develop policies regarding the use of artemisinin-based combination therapy (ACT) and strategies to eliminate malaria. K13 genotyping is integrated into surveillance activities by combining clinical efficacy studies with in vitro susceptibility testing and is supported by gene-editing studies. The validated or confirmed K13 mutations that are associated with resistance to artemisinins are Y493H, R539T, I543T, and C580Y, and the associated K13 mutations are P441L, F446I, G449A, N458Y, P553L, R561H, V568G, P574L, and A675V. The validation of a K13 mutation as a resistance marker (i.e., confirmed K13 mutation) is based on the following criteria: a significant association with delayed clearance of Plasmodium falciparum and a reduced drug sensitivity (survival rate, >1%) on ex vivo ring-stage survival assay (RSA) or in vitro RSA performed on field isolates, culture-adapted parasites, or gene-edited parasite lines, as compared with the K13 wild-type parent control. A K13 mutation is deemed to be associated with artemisinin resistance if it has a significant association with delayed clearance. A K13 mutation is said to be neutral if it has no significant association with delayed clearance or reduced drug sensitivity. A new K13 mutation is one that has been never observed and thus does not appear in mutation databases. Recommendations regarding efficacy trials and treatment policies are provided by the World Health Organization (WHO).