Previous Article | Next Article ![]()
Antimicrobial Agents and Chemotherapy, August 2009, p. 3405-3410, Vol. 53, No. 8
0066-4804/09/$08.00+0 doi:10.1128/AAC.00024-09
Copyright © 2009, American Society for Microbiology. All Rights Reserved.

Hospital for Tropical Diseases,1 HPA Malaria Reference Laboratory,2 Immunology Unit,3 Pathogen Molecular Biology Unit, Department of Infectious and Tropical Diseases, London School of Hygiene and Tropical Medicine, London, United Kingdom4
Received 7 January 2009/ Returned for modification 19 March 2009/ Accepted 3 May 2009
|
|
|---|
|
|
|---|
Quinine remains an effective regimen for treating P. falciparum malaria and is still used in the United Kingdom for the treatment of both uncomplicated and severe malaria (9). At the Hospital for Tropical Diseases (HTD), patients with P. falciparum infections are treated with quinine until parasites are undetectable in peripheral blood; patients are then given a full dose of the fixed-combination antifolate sulfadoxine-pyrimethamine (SP). The long half-lives of both active components of SP in serum provide prophylaxis against the subsequent recrudescence of parasites surviving quinine therapy, or any parasites newly emerging from hepatic schizonts after the cessation of primary therapy (10, 15). This policy appears to have provided efficacious treatment for malaria patients, but there is no active follow-up of malaria patients at present.
The efficacy of quinine does not appear to have been significantly diminished over time by the evolution of parasite resistance, although studies of in vitro susceptibility of malaria parasites show some variability in sensitivity to quinine among South American isolates (6). In contrast, SP treatment failure due to high-level parasite resistance is widespread in Asia and common in East Africa (3), although SP retains good efficacy in West Africa (4, 17). Failure of SP therapy is associated with the accumulation of point mutations in two parasite genes, pfdhfr and pfdhps, encoding the folate biosynthesis pathway enzymes dihydrofolate reductase and dihydropteroate synthetase (DHPS), respectively. The continued efficacy of SP in West Africa is probably due to the absence or rarity of some of these mutations, particularly those at codon 164 of pfdhfr (widespread in Asia) and at codon 540 of pfdhps (widely reported from East and Southern Africa). Therefore, ongoing surveillance of the geographic distribution of mutations in pfdhfr and pfdhps is of great importance for informing treatment policy both in countries where malaria is endemic and in those where it is not, such as the United Kingdom, where a large number of imported P. falciparum malaria cases are treated each year.
In March 2007, a P. falciparum malaria patient treated at the HTD with quinine plus SP returned with recrudescent malaria and was confirmed as a case of therapeutic failure. It was hypothesized that quinine, a drug rapidly cleared from host circulation, had failed to eradicate all the parasites and that subsequent administration of long-lasting SP had selected a subpopulation of parasites resistant to antifolates that survived and subsequently caused a recrudescent infection. We report here a parasitological evaluation of this particular case and a concurrent survey of molecular markers of drug resistance in parasites from 44 HTD patients presenting with confirmed P. falciparum malaria over 12 months prior to the presentation of this case and in a further 39 isolates from P. falciparum malaria cases referred from all over the United Kingdom to the Health Protection Agency (HPA) Malaria Reference Laboratory (MRL) at the London School of Hygiene and Tropical Medicine.
|
|
|---|
qPCR. Parasite DNA was extracted from 200 µl of peripheral blood using the Qiagen QIAamp blood extraction kit and was eluted in a volume of approximately 100 µl. An in-house quantitative real-time PCR (qPCR) assay (11) was employed to quantify parasite density. Primers PfalVL (GCCGAAAGGCGTAGGTAATC) and PfalVR (GTACAAAGGGCAGGGACGTA) were designed to amplify a 134-bp fragment of the 18S ribosomal subunit gene located on chromosome 5 of P. falciparum (sequence MAL5_18S, sourced from PlasmoDB [http://plasmodb.org/plasmo/; last accessed 21 April 2009]). An identical sequence is found on chromosome 7 (MAL7_18Sa). DNA (extracted as described below) was amplified in a 20-µl reaction volume, comprising 10 µl QuantiTect Sybr green mix (Qiagen, United Kingdom) and 2 µM each primer. Amplification was performed in a Rotorgene RG3000 thermocycler (Corbett, Sydney, Australia) with the following profile: 15 min at 95°C for activation of reagents, followed by 40 cycles of 94°C (30 s), 55°C (40 s), and 68°C (50 s). Products were stabilized at 68°C for 5 min before the following melt analysis was performed: 55°C for 45 s, followed by a temperature increase of 0.5°C per step with 5 s at each step of the gradient. Fluorescent data were collected on the 6-carboxyfluorescein/Sybr channel of the RG3000 thermocycler and were quantified against a serial dilution of the recently described WHO international standard for P. falciparum DNA (7).
Determination of parasite markers of drug resistance. Resistance to the antimalarial chloroquine (CQ) is associated with the presence of mutations in the pfcrt gene, encoding the P. falciparum CQ resistance transporter (2). Alleles associated with resistance to CQ were identified by real-time double-labeled hydrolysis probe PCR in a Corbett RG3000 thermocycler, as described previously (13).
Allele sequences of pfdhfr were identified using a PCR amplification and direct-sequencing protocol designed specifically for this study. Briefly, 5 µl of DNA was subjected to PCR with forward primer FBR01 (AAGCAAAAATGAGGGGAAAAA) and reverse primer FBR02 (ACATCGCTAACAGAAATAATTTGA). The DNA was amplified under standard conditions with the following cycling program: 93°C for 5 min; 40 cycles of 93°C for 30 s, 56°C for 30 s, and 68°C for 75 s; and 68°C for 5 min.
Amplification products were purified using Qiagen PCR Elute minicolumns, checked by agarose electrophoresis, and then sequenced in both directions using the ABI BigDye (version 3.1) sequencing kit according to the manufacturer's instructions, but all volumes were scaled down eightfold. Sequencing products were purified by alcohol precipitation and fractionated on an ABI 3770 sequencer. Chromatogram data were analyzed and manually curated using Chromas software. Good-quality DNA sequences from both strands were required in order for any sequence to contribute to analysis.
Allele sequences of pfdhps were amplified using the nested-PCR procedure of Pearce et al. (8) as described by these authors. PCR products were sequenced as for pfdhfr by using the following sequencing primers: AACCTAAACGTGCTGTTCAA (forward), AATTATTAAAAAAAAAAAAC (forward), AATTGTGTGATTTGTCCACAA (reverse), and TTTTAATAATTTTATAATTTC (reverse).
Estimating the multiplicity of infection.
The multiplicities of infection of different isolates from the index case were estimated using two polymorphic sequence tags with variable-length trinucleotide repeats, poly
on P. falciparum chromosome 4 and PfPK2 on chromosome 12, as previously described (1). Neither of these markers is closely linked to genetic loci associated with antifolate resistance; pfdhfr is located on chromosome 4 (220 kb from poly
), and pfdhps is on chromosome 8.
Statistical analysis was performed in Stata, version 8.0 (StataCorp, College Station, TX). Categorical variables were tested for associations by using the two-sided Fisher exact test.
|
|
|---|
A 37-year-old Caucasian male presented to the HTD as an outpatient on 21 February 2007 with a 3-day history of myalgia and a 1-day history of lethargy, headache, and fever. He had been traveling in Mozambique for 1 month, staying at a backpackers' hostel in a coastal resort, and had not used any malaria prophylaxis. He returned to the United Kingdom on 16 February 2007, and his first symptoms began 2 days later.
On examination, the patient was not acutely unwell; he was conscious and alert, with no signs of meningism. His temperature was elevated (39.9°C); there was no jaundice or lymphadenopathy; and his chest was clear. Sinus tachycardia was observed, but cardiac indicators were otherwise normal. There was no evidence of hepatosplenomegaly or rash. The hemoglobin level was normal, at 16.0 g/dl; the leukocyte count was 4.9 x 109 cells per liter; and the platelet count was 141 x 109/liter. Urea, electrolytes, and liver function tests were normal. Microscopy of a Giemsa-stained blood film revealed the presence of P. falciparum ring-stage trophozoites in 2.1% of red blood cells.
The patient was admitted to hospital and treated with intravenous quinine (one dose) followed by oral quinine (eight doses). Thick and thin blood films were examined daily for malaria parasites; asexual stages of P. falciparum were undetectable by microscopy after 3 days of treatment. The patient was given a single dose of SP and discharged the following day (26 February).
On 13 March 2007, 15 days after discharge, the patient returned to the HTD with fever, anorexia, and myalgia. He presented with a temperature of 38.7°C. A malaria blood film revealed late trophozoites of P. falciparum infecting 0.6% of erythrocytes. The patient was readmitted to the HTD as an inpatient and treated with AP daily for 3 days. He made a good recovery. On follow-up 2 months later, the patient reported being well, with no recurrence of symptoms.
Parasitological history. A total of 10 peripheral blood samples were taken from the patient for primary diagnosis and for monitoring of parasite clearance under treatment: 5 samples from the quinine- and SP-treated episode and 5 samples from the AP-treated episode. To test whether the rates of parasite clearance differed between the quinine- and AP-treated episodes, we used qPCR to estimate parasite clearance rates for the two episodes. The relative parasite density (measured against a serial dilution of P. falciparum-positive control DNA from a previous patient) was plotted over time for both episodes, with time zero representing the times of the first blood draws on 21 February and 13 March, respectively. The data are plotted against the newly established WHO international standard for P. falciparum DNA (7) in Fig. 1. Ninety percent parasite clearance (16) was estimated to have occurred by about 25 h under quinine treatment and 59 h under AP treatment. However, our sequential parasitological samples were taken 18 to 24 h apart, and the precise time of sampling was not always recorded. Therefore, although our data suggest that parasitological clearance was significantly faster under quinine than under AP therapy, more closely spaced and more accurately timed sampling would be required to verify this observation.
![]() View larger version (12K): [in a new window] |
FIG. 1. Quantitative PCR analysis of two sequential P. falciparum malaria episodes for the same patient. Episode 1 was treated with quinine; episode 2, with AP. Each point represents the mean estimate of parasite density from two replicates in a single experiment. The y axis shows parasite density, expressed in kilo-international units of P. falciparum DNA, on a logarithmic scale. The x axis shows the approximate time (in hours) after the first diagnostic blood sample was taken. Where the exact time was not recorded, an interval of 24 h between consecutive samples is assumed.
|
and PfPK2 markers, which indicated that at the time of first presentation in February 2007, our primary case was an infection with a minimum of four distinct parasite clones, but that these were reduced to one clone following treatment during the second episode. |
View this table: [in a new window] |
TABLE 1. Evaluation of resistance-associated markers in two malaria episodes in the same patient
|
|
View this table: [in a new window] |
TABLE 2. pfcrt and pfdhfr genotypes among P. falciparum isolates from malaria patients presenting to the HTD in 2006
|
|
View larger version (19K): [in a new window] |
FIG. 2. Alignment of deduced PfDHPS amino acid sequences from 46 P. falciparum isolates. The eight different PfDHPS haplotypes encoded by DNA sequences determined in this study from patients at the HTD in 2006 are shown, and the number of isolates with each haplotype is given on the left (#). Haplotypes are arranged in increasing order of amino acid substitutions, and substitutions relative to the wild-type sequence are shaded. Amino acids 445 to 530 were invariant and are omitted for clarity. Amino acids 1 to 424 were not sequenced, because there are no previous reports of amino acid substitutions in this region of the polypeptide. Where mixed infections were present, only the dominant haplotype is shown and contributes to the number of isolates given. The novel genotypes VAGKAA and VAGKGS are shown in boldface.
|
|
View this table: [in a new window] |
TABLE 3. pfdhps haplotypes and countries of origin for 46 P. falciparum isolates from malaria patients presenting to the HTD, London, United Kingdom, in 2006
|
|
View this table: [in a new window] |
TABLE 4. pfdhps haplotypes and countries of origin for 39 P. falciparum isolates of West African origin from malaria patients presenting across the United Kingdom in 2007
|
|
|
|---|
Our findings are consistent with previous work showing that, in East Africa, the pfdhps K540E mutation is closely linked to SP treatment failure, specifically by disrupting the antiparasite effects of the sulfadoxine component of the drug combination (3). This mutation is absent or rare in West Africa, where SP efficacy remains good (5). The HTD malaria patient base largely comprises infections picked up in West Africa, and by combining the results from our survey of pfdhps haplotypes among these patients with those from the 39 additional West African isolates from the MRL, the odds ratio of a malaria case imported to the United Kingdom from West Africa, rather than from East or Southern Africa, carrying the K540E mutation is 0.0041 (95% confidence interval, 0.0001 to 0.065; P < 0.0001). This is consistent with observations in the field across East, West, and Southern Africa, which have recently been summarized in highly functional graphic maps (http://www.lshtm.ac.uk/pmbu/drm/map-540.html).
The observation among Nigerian isolates of a previously undescribed amino acid substitution, valine for isoleucine at codon 431, was unexpected, and its significance remains unclear. Its association with the relatively uncommon substitutions A581G and A613S in 7 of the 10 isolates found is consistent with a recent origin of the codon 431 substitution in a parasite carrying the AGKGS haplotype. Alternatively, functional constraints on the enzyme itself may limit the ability of this mutation to combine with some other haplotypes. The role, if any, of this mutation in parasite resistance to SP needs to be determined, and further work in Nigeria and neighboring countries to examine the sensitivity of this parasite genotype to a variety of antifolate drugs is required. One intriguing possibility is that widespread use of the antibiotic combination trimethoprim-sulfamethoxazole as prophylaxis against Pneumocystis jirovecii pneumonia for people infected with human immunodeficiency virus in Africa, particularly pregnant women, may be placing additional sulfonamide selective pressure for new alleles at the pfdhps locus that are not directly related to antimalarial use.
In conclusion, the treatment of imported cases of uncomplicated P. falciparum malaria with quinine and SP is threatened by the large proportion of such infections harboring mutations in pfdhfr and pfdhps. Our data suggest that the K540E mutation in pfdhps may be particularly important, although currently this mutation is common only among cases imported from East Africa. Surveillance for further spread of this genotype into West Africa, as indicated by its occurrence in two Ghanaian isolates, is required. The antifolate sensitivity profiles of the novel pfdhps haplotypes identified here remain to be determined.
We thank Cally Roper for useful discussions and for suggesting the microsatellite analysis.
All the authors declare no conflict of interest.
Published ahead of print on 11 May 2009. ![]()
|
|
|---|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2010 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»