Resistance to antifolates in Plasmodium falciparum monitored by sequence analysis of dihydropteroate synthetase and dihydrofolate reductase alleles in a large number of field samples of diverse origins

https://doi.org/10.1016/S0166-6851(97)00114-XGet rights and content

Abstract

Resistance of Plasmodium falciparum to antifolate chemotherapy is a significant problem where combinations such as Fansidar (pyrimethamine–sulfadoxine; PYR–SDX) are used in the treatment of chloroquine-resistant malaria. Antifolate resistance has been associated with variant sequences of dihydrofolate reductase (DHFR) and dihydropteroate synthetase (DHPS), the targets of PYR and SDX respectively. However, while the nature and distribution of mutations in the dhfr gene are well established, this is not yet the case for dhps. We have thus examined by DNA sequence analysis 141 field samples from several geographical regions with differing Fansidar usage (West and East Africa, the Middle East and Viet Nam) to establish a database of the frequency and repertoire of dhps mutations, which were found in 60% of the samples. We have also simultaneously determined from all samples their dhfr sequences, to better understand the relationship of both types of mutation to Fansidar resistance. Whilst the distribution of mutations was quite different across the regions surveyed, it broadly mirrored our understanding of relative Fansidar usage. In samples taken from individual patients before and after drug treatment, we found an association between the more highly mutated forms of dhps and/or dhfr and parasites that were not cleared by antifolate therapy. We also report a novel mutation in a Pakistani sample at position 16 of DHFR (A16S) that is combined with the familiar C59R mutation, but is wild-type at position 108. This is the first observation in a field sample of a mutant dhfr allele where the 108 codon is unchanged.

Introduction

The combination of pyrimethamine (PYR) and sulfadoxine (SDX), known as Fansidar or PSD, is a cheap and effective agent against chloroquine-resistant falciparum malaria that is in widespread use in Africa and other parts of the world. However, resistance to this formulation, long established in parts of South East Asia, now threatens to leave Africa with no treatment affordable on a mass scale. The detailed molecular basis of Fansidar failure is as yet unclear. Variant sequences of Plasmodium falciparum dihydrofolate reductase (DHFR), the target enzyme of PYR, were first described in 1988, and it is now well established that high level PYR resistance results from the accumulation of mutations in the dhfr gene, principally at codons 108, 59 and 51 (reviewed in [1]). Conclusive experimental proof of the importance of these mutations was provided by transformation of wild-type parasites with constructs bearing mutant forms of dhfr [2]. More recently, mutations have also been reported in the gene encoding dihydropteroate synthetase (DHPS), the target of SDX 3, 4. In addition, the progeny of a genetic cross between two cloned lines, together with a number of unrelated lines, were used to demonstrate that the inheritance of mutations in the dhps gene conferred upon the progeny a similar level of SDX resistance to that shown by the mutant parent, and that the degree of resistance broadly correlated with the number of altered residues in DHPS [5]. These studies also enabled the wild-type sequence of DHPS to be defined. The most SDX-resistant parasites that have been assayed in vitro carry mutations affecting positions 436 (Ser to Phe; S436F), 437 (Ala to Gly; A437G) and 613 (Ala to Ser; A613S), and display IC50 values (drug concentration inhibiting growth to 50% of control levels) ca. 3 orders of magnitude higher than parasites carrying the wild-type DHPS enzyme [5].

The existence of a number of allelic variants of dhps that confer SDX resistance raises a major question: how do such mutations contribute to Fansidar failure, relative to the mutations in dhfr that confer PYR resistance? To date, 10 different mutant genotypes for dhfr have been reported from large numbers of field samples (e.g. 6, 7, 8), and in the initial studies to characterise the dhps gene, 7 different genotypes affecting 4 codons were reported from a total of 15 laboratory-cultured parasite lines 3, 4. It is now important to investigate whether further variants of the DHPS enzyme exist in field samples, to identify what combinations of DHPS and DHFR sequences occur together, to determine the relative frequency and distribution of such combinations, and to establish the minimum requirements for parasites to become capable of overcoming Fansidar inhibition.

Whereas mutation in dhfr appears to be the major, if not the sole contributor to PYR resistance, the role of the dhps mutations in determining levels of SDX resistance is less clearly defined. The antagonism to SDX inhibition exhibited by exogenous folate and other compounds related to SDX, such as p-aminobenzoic acid and p-aminobenzoylglutamate, has long been known 9, 10, 11, and it has been recently demonstrated that different parasite lines can vary greatly in their susceptibility to this effect in vitro 5, 12. These studies also showed that concentrations of folate corresponding to those found in human plasma (though in a different oxidation state) were capable, in some instances, of conferring levels of SDX resistance that were even higher than those arising from the triply mutated dhps genes. However, the relative importance of dhps mutations and folate antagonism during Fansidar challenge in vivo is still unclear, since the presence of PYR in the drug combination is an added factor that also influences folate antagonism of SDX sensitivity (P. Wang and J. E. Hyde, unpublished).

In this work, we analyse dhps/dhfr combinations in several collections of P. falciparum field samples from different regions of Africa, the Middle East and Viet Nam. These represent areas with a wide variation in previous and current usage of Fansidar. All of these samples contributed to a large population study, where we established the incidence of mutation and degree of variation found in the dhps sequences. With the samples from East Africa, we were additionally able to examine the relationship between antifolate drug efficacy and dhps/dhfr sequences, as these sets included samples taken before and after treatment of a particular patient. All samples have been analysed by DNA sequencing, rather than by mutation-specific PCR 8, 13, 14, 15, 16, as sequencing reveals all variants in all known mutant positions, and is capable of picking up previously unreported mutations, as observed below.

Section snippets

Extraction of DNA from blood samples

Infected blood was collected as finger-prick samples air-dried onto a piece of filter paper. Squares of about 5 mm2 excised from the original were soaked in 0.5 ml phosphate-buffered saline (PBS) containing 0.05% saponin at room temperature for 30 min, then washed once in 0.5 ml PBS. After adding 150 μl 5% Chelex suspension, the sample was boiled for 10–15 min, then spun down. The supernatant was transferred to a fresh tube and stored at −20°C until use. Depending upon the type of filter paper,

Results

The samples we obtained from the Middle East, Mali and Viet Nam (Section 3.1–Section 3.3) were one-time collections from patients with parasitaemia, taken before drug treatment. These samples thus represent a population survey where variants in the dhps and dhfr sequences of P. falciparum are determined simultaneously. The samples from Kenya and Tanzania, however (Section 3.4–Section 3.5), were taken from patients both before and after antifolate drug treatment. Thus, in addition to making a

Discussion

Earlier work 3, 5has demonstrated that, as the number of mutations affecting positions 436, 437, 581 and 613 of DHPS accumulate, the susceptibility of P. falciparum to SDX inhibition is markedly reduced by up to 3 orders of magnitude. It has also been shown that these mutations, as well as the K540E mutation observed initially in several Thai samples, result in reduced levels of drug binding to the target enzyme [22]. The analysis presented here directly addresses the question of the relative

Acknowledgements

We thank Joe Cortese and Chris Plowe (University of Maryland), for material assistance and helpful discussions, F. Baidas, N. Hamo and M. Ibrahim (Al-Ain) for collection of samples in the UAE, NIH/Mali-Tulane TMRC and TDR/WHO RSG for support of sample collection in Mali, the Swedish Agency for Research Cooperation with Developing Countries (sample collection in Viet Nam), the Swiss Agency for Development and Cooperation (sample collection in Tanzania), the Medical Research Coordinating

References (27)

  • T. Triglia et al.

    Primary structure and expression of the dihydropteroate synthetase gene of Plasmodium falciparum

    Proc Natl Acad Sci USA

    (1994)
  • P. Wang et al.

    Sulfadoxine resistance in the human malaria parasite Plasmodium falciparum is determined by mutations in dihydropteroate synthetase and an additional factor associated with folate utilisation

    Mol Microbiol

    (1997)
  • Plowe CV, Cortese JF, Djimde A et al. Mutations in Plasmodium falciparum dihydrofolate reductase and dihydropteroate...
  • Cited by (0)

    View full text