| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
 |
ABSTRACT |
 TOP ABSTRACT INTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
119 nM, and six isolates had low IC50s (five wild-type or mixed dhfr, 0.04–1.37 nM; one triple mutant, 6.4 nM). Of 194 isolates, only 7 had the wild-type dhfr and 187 were mutants. The results of the two methods were highly concordant and indicated a significant increase (P < 0.05) in the prevalence of mutant, pyrimethamine-resistant P. falciparum between 1994 and 2005. The addition of probenecid or sulfinpyrazone to pyrimethamine resulted in a slight-to-moderate decrease in the level of in vitro pyrimethamine resistance without rendering the parasites susceptible to pyrimethamine. Analysis of molecular markers may be useful for the long-term surveillance of antifolate-resistant malaria. |
INTRODUCTION |
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The massive use of antifolate drugs leads to a rapid development of drug resistance, as it occurred in southeast Asia.4 Therefore, sulfadoxine-pyrimethamine monotherapy is not expected to remain highly effective in Africa in the future unless novel therapeutic strategies are applied in the field and target populations for the use of sulfadoxine-pyrimethamine are well defined. Treatment failures after sulfadoxine-pyrimethamine monotherapy have already been reported from many areas in Africa.5 In this context, it is important to set up a surveillance system for a rapid and appropriate response to the changing epidemiology of antifolate-resistant P. falciparum. In vitro drug susceptibility assay and molecular analysis of dihydrofolate reductase (dhfr) and dihydropteroate synthase (dhps), targets of pyrimethamine and sulfadoxine, respectively, are alternative epidemiologic tools to describe the evolution of antifolate resistance. In vitro pyrimethamine resistance is strongly associated with a single Ser108Asn substitution in DHFR, and additional substitutions (Asn51Ile, Cys59Arg, and Ile164Leu) increase the level of resistance.6–8 In vitro sulfadoxine resistance is associated with amino acid changes occurring at positions 436, 437, 540, 581, and 613 in DHPS.9 The first objective of the present study was to determine the dhfr sequence and in vitro response to pyrimethamine of clinical isolates obtained in 2004–2005 and compare the results with those of the preceding years to analyze the dynamics of the spread of pyrimethamine-resistant P. falciparum in Yaoundé, Cameroon.
One of the experimental strategies to circumvent drug resistance is to combine an antimalarial drug with a pharmacological agent (calcium channel blockers, tricyclic antidepressants, tricyclic antihistamines, phenothiazines, and others), collectively called a modulator, which modifies drug transport and increases the intracellular concentration of the antimalarial drug.10 Methotrexate is a DHFR inhibitor commonly used as an anticancer agent. One of the proposed mechanisms of methotrexate resistance in human cancer cells is an enhanced drug efflux mediated by multidrug resistance protein 1 (MRP1), which is an ATP-binding cassette drug transporter.11,12 Methotrexate resistance can be modulated in vitro in mammalian cells by probenecid and sulfinpyrazone.13 Based on these observations, Nzila and others14 have shown that the addition of probenecid (50 µM) results in a sevenfold increase in pyrimethamine susceptibility in a highly resistant laboratory-adapted P. falciparum strain. If this experimental approach is validated, pyrimethamine resistance may be circumvented. The second objective of the present study was to evaluate the effectiveness of this experimental approach in clinical isolates collected in the field.
|
PATIENTS AND METHODS |
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
12 years of age and adults seeking consultation at the Nlongkak Catholic Missionary Dispensary in Yaoundé, Cameroon during 2004–2005 after informed consent was obtained. The following criteria were used: the presence of P. falciparum at a parasitemia 0.1% and a negative Saker-Solomons urine test result for 4-aminoquinolines.15 Patients with severe anemia (hematocrit < 15%), pregnant women, and those with signs and symptoms of severe and complicated malaria were excluded. All enrolled patients were treated with amodiaquine, amodiaquine–sulfadoxine-pyrimethamine combination or an artesunate-amodiaquine combination. The study was reviewed and approved by the Cameroonian National Ethics Committee and the Cameroonian Ministry of Public Health.
In vitro drug assay.
Venous blood samples were collected and transported to our laboratory within two hours after venipuncture. Blood samples were washed three times by centrifugation with p-aminobenzoic acid (PABA)– and folic acid–free RPMI 1640 culture medium (Sigma Chemical Co., St. Louis, MO). Washed erythrocytes were resuspended in complete PABA- and folic acid–free RPMI 1640 medium containing buffers (25 mM HEPES and 25 mM NaHCO3) and 10% fetal bovine serum (batch no. 5–41201; Integro b. v., Amsterdam, The Netherlands) at a hematocrit of 1.5%. The initial parasitemia was adjusted to 0.6% by the addition of uninfected erythrocytes if the parasitemia was 1.0%.
Pyrimethamine, probenecid, and sulfinpyrazone were obtained from the Sigma Chemical Co. Stock solutions and dilutions of pyrimethamine, probenecid, and sulfinpyrazone were prepared in absolute ethanol. Drug solutions were distributed in duplicate in 96-well culture plates and dried. The final concentrations of pyrimethamine, probenecid, and sulfinpyrazone ranged from 0.0488 nM to 51,200 nM (in fourfold dilutions), 6.25 to 6,400 µM (in two-fold dilutions), and 0.625 to 640 µM (in two-fold dilutions), respectively. Fixed concentrations of probenecid (25 µM or 50 µM) or sulfin-pyrazone (12.8 µM) were added to the pyrimethamine assay plate and dried.
In vitro isotopic drug assays were performed as previously described.16,17 Two hundred microliters of the suspension of infected erythrocytes were distributed in each well of the 96-well tissue culture plates. The parasites were incubated at 37°C in an atmosphere of 5% CO2 for 42 hours. 3H-hypoxanthine (1 µCi/well; Amersham International, Little Chalfont, United Kingdom) was added at the beginning of the incubation period to measure parasite growth. The plates were frozen to terminate the assay. The plates were thawed, and the contents of each well were collected on glass-fiber filter papers, washed, and dried using a cell harvester. The filter disks were transferred into scintillation vials, and 2 mL of scintillation cocktail (Organic Counting Scintillant® Amersham International) were added. The incorporation of 3H-hypoxanthine was quantitated using a liquid scintillation counter (Wallac 1409; Pharmacia, Uppsala, Sweden). The 50% inhibitory concentration (IC50), defined as the drug concentration at which 50% of the incorporation of 3H-hypoxanthine is inhibited compared with that of drug-free control wells, was determined by nonlinear regression analysis using the PrismTM software (GraphPad Software, Inc., San Diego, CA).
Polymerase chain reaction (PCR) and DNA sequencing. Venous blood (50 µL) was imbibed onto Isocode Stix® filter papers (Schleicher and Schuell, Ecquevilly, France). The filter papers were dried overnight in an incubator at 37°C and stored at –20°C in airtight plastic bags until analysis. Parasite DNA was extracted from filter papers by the boiling method.18 Briefly, after rinsing once in 500 µL of sterile distilled water, the filter paper was placed into a 0.5-mL microtube to which 75 µL of sterile distilled water was added. The filter paper was incubated at 100°C for 20 minutes and agitated for a few seconds. The supernatant (10 µL) was used directly for amplification of dhfr gene fragment.
The entire DHFR domain (708 basepairs) of the dhfr-thymidylate synthase (dhfr-ts) gene was initially amplified by the primary PCR in a mixture consisting of genomic DNA (10 µL of supernatant containing parasite and human DNA), 15 pmol of a pair of synthetic oligonucleotides (forward primer ST1L, 5'-TGATGGAACAAGTCTGCGACGTTTTC-GAT-3' and reverse primer ST2L, 5'-TTCATTTAACATTT-ATTATTCGTTTTCTT-3'), buffer (50 mM KCl, 10 mM Tris, pH 8.3), 1.5 mM MgCl2, 200 µM of deoxynucleoside triphosphates (mixture of dGTP, dATP, dTTP, and dCTP), and one unit of Taq DNA polymerase (Roche Diagnostics, Meylan, France) in a total volume of 50 µL. The PTC-100 thermal cycler (MJ Research, Watertown, MA) was programmed as follows: 94°C for 2 minutes in the first cycle and 1 minute in subsequent cycles, 50°C for 90 seconds in the first cycle and 1 minute in subsequent cycles, and 72°C for 1 minute in all cycles, for a total of 30 cycles. In the secondary nested PCR, a 499-basepair fragment was amplified from the primary amplification product (0.75–1 µL) by using the primer pairs DHFR-31 (forward primer, 5'-ATTT-ATGCCATATGTGCATGTTGTAAC-3') and DHFR-529R (reverse primer, 5'-CTTTTCTAAAAATTCTTGAT-AAACAAC-3') in a 100-µL reaction mixture. The same thermal cycler program was used. Five microliters of the final amplification product was loaded on a 1.5% agarose gel, subjected to electrophoresis, stained with ethidium bromide, and visualized under ultraviolet transillumination to confirm the presence of the 499-basepair DNA fragment. The PCR products were sequenced using the ABI Prism automated DNA sequencer (Perkin Elmer Corp., Les Ulis, France).
Data analysis. Electropherogram data were viewed and analyzed using the Editview sequence analysis software (Perkin Elmer Corp.). An isolate was considered to be wild-type if the following haplotype was observed: Asn-51, Cys-59, Ser-108, and Ile-164.6,7 An isolate was considered to be mutant if at least one of these codons was mutated: Asn- to Ile-51, Cys-to Arg-59, Ser- to either Asn-108 or Thr-108, and Ile- to Leu-164. Previous studies have shown that Ser- to Thr-108 change occurs with Ala- to Val-16 substitution.6,7 The use of the internal primer DHFR-31, which includes codon 16, did not enable determination of the sequence of this codon in the present study. However, our previous studies using semi-nested PCR showed that none of the Cameroonian isolates have the mutant Val-16 codon.17,19
As in our previous study, the in vitro threshold for pyrimethamine resistance was arbitrarily fixed at 100 nM.17 This threshold was estimated from the results of the assays performed with RPMI–10% human serum. The use of RPMI–10% fetal bovine serum, as in the present study, results in an average three-fold decrease in the pyrimethamineIC50.20 The adjusted threshold for pyrimethamine assays using RPMI–10% fetal bovine serum is expected to be 
30 nM. The proportions of pyrimethamine resistance, based on either an IC50 100 nM (or 30 nM) or the presence of Asn-108 mutant allele, at different time periods were compared by the chi-square test for trend. Isolates with mixed dhfr alleles were not included in the statistical analysis because their in vitro response is unpredictable. The significance level (P) was fixed at 0.05.
|
RESULTS |
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Of 109 isolates with interpretable assays, most (n = 103, 94%) had an IC50 119 nM, indicating in vitro pyrimethamine resistance (Figure 1
). The geometric mean pyrimethamine IC50 (95% confidence intervals) of the resistant isolates was 1,980 nM (1,580–2,480 nM). All 103 isolates were either pure double (Arg-59/Asn-108, n = 5) or pure triple (Ile-51/Arg-59/Asn-108, n = 97) dhfr mutants or mixed isolates with Ile-51/Arg-59/Ser- and Asn-108 (n = 1). Six isolates with low IC50s had the following values: 0.04 nM (extrapolated value due to an almost complete growth inhibition at the lowest drug concentration used; wild-type dhfr), 0.19 nM (wild-type dhfr), 0.42 nM (mixed dhfr alleles with Asn- and Ile-51/Cys- and Arg-59/Ser- and Asn-108), 0.49 nM (wild-type dhfr), 1.37 nM (wild-type dhfr), and 6.4 nM (triple dhfr mutant).
|
). Initial experiments in four isolates cultivated in RPMI–human serum with three fixed concentrations of probenecid (5 µM, 25 µM, and 50 µM) added to the pyrimethamine assay plate showed slight to no effect with 5 µM probenecid and 12–36% decrease in pyrimethamine IC50 with 25 µM probenecid. Subsequent experiments were performed with isolates cultivated in RPMI–fetal bovine serum with either 25 µM or 50 µM probenecid added to the pyrimethamine plate. A mean decrease of 19% (maximum µ 73%) in pyrimethamine IC50 was obtained in 25 isolates by adding 25 The intrinsic antimalarial activity of probenecid was determined in 19 isolates using RPMI–10% fetal bovine serum.The geometric mean IC50 (range) was 53.8 µ M (3.19–273 µM) (Table 1). Initial experiments in four isolates cultivated in RPMI–human serum with three fixed concentrations of probenecid (5 µM, 25 µM, and 50 µM) added to the pyrimethamine assay plate showed slight to no effect with 5 µM probenecid. In five isolates tested with 50 The intrinsic antimalarial activity of probenecid was determined in 19 isolates using RPMI–10% fetal bovine serum.The geometric mean IC50 (range) was 53.8 µ M (3.19–273 µM) (Table 1). Initial experiments in four isolates cultivated in RPMI–human serum with three fixed concentrations of probenecid (5 µM, 25 µM, and 50 µM) added to the pyrimethamine assay plate showed slight to no effect with 5 µM probenecid, the mean decrease in pyrimethamine IC50 was 23%. However, in our experimental conditions, the fixed concentration of 50 µM seemed too close, and even above, to the intrinsic activity of probenecid for some isolates. The modulator activity of probenecid was effective in rendering a pyrimethamine-resistant isolate (assuming a cut-off 100 nM) pyrimethamine-susceptible in only three isolates (06/04, 12/04, and 17/04), all of which had a low level in vitro pyrimethamine resistance (IC50s = 119–236 nM). Experiments on pyrimethamine combined with a single fixed concentration (12.8 µM) of sulfinpyrazone showed a decrease in pyrimethamine IC50 in a few isolates (Table 2). However, the effect was inconsistent in other isolates. These experiments were not pursued after obtaining the results from 11 isolates.
|
|
Although RPMI–10% fetal bovine serum lowers pyrimethamine IC50s compared with RPMI–10% human serum,the responses of the isolates studied in 2004–2005 can be classified as 119 nM (n = 103) and 6.4 nM (n = 6). The isolates in the former group were all dhfr mutants or mixed alleles, and those in the latter group were wild-type or mixed alleles, with the exception of one isolate (IC50 = 6.4 nM) that had triple mutations. Based on this observation, the arbitrary criterion used to distinguish between pyrimethamine-susceptible isolates (IC50< 100 nM) and pyrimethamine- resistant isolates (IC50 100 nM) in our previous studies was also applied in the present study for comparison with our results from 1994 to 2001.22–24 The evolution of pyrimethamine resistance in Yaoundé between 1994 and 2005 (n = 401 for in vitro assays and 516 for dhfr) showed an increasing prevalence (P < 0.05) of pyrimethamine-resistant and dhfr-mutant P. falciparum isolates (Figure 2). The baseline prevalence of pyrimethamine resistance determined in Yaoundé in 1994–1995 was 41% and 43% (48% including mixed alleles) using in vitro assay and PCR, respectively. The prevalence of pyrimethamine resistance increased steadily during the past 12 years to attain 94% and 95% (96% including mixed alleles) using in vitro assay and PCR, respectively.
|
|
DISCUSSION |
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Our previous studies, as well as the present study, have demonstrated the high correlation between in vitro pyrimethamine assay and molecular analysis of dhfr.22–24 This finding implies that the two methods do not need to be performed in parallel for further monitoring. In vitro assay is more labor-intensive and more cumbersome to perform in aseptic conditions in the field. It also requires expensive equipment and is best performed in a well-equipped laboratory. The successful rate of interpretable assays may be low(< 80%), especially in populations in whom self-medication is a common practice. Moreover, in vitro assay for pyrimethamine is difficult to standardize because of the interference of PABA and folic acid, which are present in the standard RPMI 1640 medium, human serum, and erythrocytes, with drug action. In contrast, a large number of finger prick capillary blood samples can be collected on filter papers with more ease and more rapidly in the field. The success rate of PCR is usually very high (> 95%), even with samples from patients who have self-medicated before consultation and from patients with low parasitemia (< 0.1%). A high output nested PCR can amplify the target gene of up to 96 samples simultaneously in a single PCR plate in one day. These technical advantages indicate that further surveillance of pyrimethamine resistance in Cameroon can be entirely based on molecular studies.
Molecular surveillance of antifolate resistance should include the search for triple dhfr mutations, which are a necessary but insufficient condition for clinical resistance to sulfadoxine-pyrimethamine monotherapy, and the appearance of mutations that are associated with a high level of pyrimethamine resistance, such as Ile164Leu,27 but are not presently encountered in central Africa. Furthermore, because pyrimethamine (or biguanides such as proguanil and chlor-proguanil) is always administered with either sulfones or sulfonamides, a concomitant search for amino acid substitutions in DHPS, in particular Ala437Gly and Lys540Glu, which with the triple dhfr mutations (called quintuple mutations) are associated with therapeutic failure to sulfadoxine-pyrimethamine in east Africa, is required to provide complementary information for epidemiologic studies and for monitoring sulfadoxine-pyrimethamine resistance within the context of intermittent preventive treatment of pregnant women in Africa.28
Our series of studies using two complementary methodologic approaches22–24 have shown an increasing prevalence of pyrimethamine resistance in Yaoundé since 1994. The increasing prevalence is largely explained by the steady replacement of wild-type isolates (52% in 1994–1995 and only 3.6% in 2004–2005) by triple dhfr mutants during the 12-year period in Yaoundé. Single and double dhfr mutants constitute only a few isolates. Longitudinal and cross-sectional studies at other sites in Cameroon showed a similar pattern of dhfr allelic distribution.19
A model that may explain this phenomenon is an introduction into Cameroon of an original triple mutant P. falciparum, possibly from southeast Asia via east Africa in the late 1980s or early 1990s,29 and the subsequent genetic transmission of triple dhfr mutations to new generations of parasites that were under increasing drug pressure. This transmission is probably non-clonal because of the high rate of multiple parasite populations within a single clinical isolate, which favors recombination during the sporogonic phase in mosquitoes, and a wide polymorphism of the P. falciparum isolates circulating in southern Cameroon.30 A single or double mutant, or mixed P. falciparum co-existing with the triple mutant, may have been diluted within a large parasite population of triple dhfr mutants.
An alternative model is an independent lineage of mutant parasites arising from either single, double, or triple mutations occurring at once or a gradual acquisition and stepwise accumulation of dhfr mutations (from a single Ser108Asn replacement to one of the alternative double substitutions [Asn51Ile–Ser108Asn or Cys59Arg–Ser108Asn], then to triple [Asn51Ile–Cys59Arg–Ser108Asn] dhfr mutations) by the local P. falciparum parasite in central Africa. The latter hypothesis of multiple, independent origins of dhfr mutant P. falciparum is supported by one recent study based on the microsatellite analysis of east African malaria parasites.31 These two models are not necessarily contradictory.
The results of other microsatellite studies support the epidemiologic model of regional expansion of dhfr mutant P. falciparum at the original foci of antifolate-resistant malaria (southeast Asia and the Amazon basin).32,33 Further studies using microsatellite markers would be required to advance an evidence-based hypothesis that may explain the spread of antifolate-resistant malaria parasites in Cameroon and to determine whether the original dhfr mutant P. falciparum was introduced from outside or emerged locally.
It cannot be predicted whether antifolate drugs will regain their clinical efficacy if they were entirely withdrawn for several years from a stable malaria-endemic area. Such a scenario would be unlikely because sulfadoxine-pyrimethamine is officially recommended for the intermittent preventive treatment of pregnant women and generic and counterfeit drugs are widely circulating through unofficial outlets in central Africa.34 One possible strategy would involve the use of a modulator to enhance the efficacy of pyrimethamine. Based on the initial in vitro findings of the modulator effect of probenecid on pyrimethamine,14 several clinical studies were conducted in Nigeria to assess the efficacy of sulfadoxine-pyrimethamine-probenecid combination.35–37 This combination was reported to be well-tolerated and more effective than sulfadoxine-pyrimethamine alone, but probenecid had no effect on sulfadoxine-pyrimethamine–induced gametocytogenesis. Although previous experimental data suggested that probenecid decreases folate uptake into malaria-infected erythrocytes, the exact underlying mechanism of action of sulfadoxine-pyrimethamine–probenecid combination in malaria parasites is unknown.14 In our present in vitro study, the modulator effect of probenecid on pyrimethamine activity seemed inadequate to reverse pyrimethamine resistance. Our results may suggest that altered drug transport may not be the major mechanism of resistance involved or that in vitro culture in PABA- and folic acid–deficient medium may alter drug and/or substrate transport. Further in vitro experiments are required to elucidate whether a homolog of human MRP1 exists in P. falciparum and how probenecid enhances pyrimethamine activity.
The present longitudinal study demonstrates the rapid replacement of wild-type P. falciparum by triple dhfr mutants in Yaoundé over a 12-year period. The increasing prevalence of dhfr mutants is strongly supported by in vitro drug susceptibility assays performed in parallel with PCR and DNA sequencing. Based on the evaluation of clinical efficacy during the 14-day follow-up period, sulfadoxine-pyrimethamine monotherapy remained moderately effective in the clinical studies conducted between 1999 and 2004.25 The prevalence of in vitro pyrimethamine resistance increased during the same period.24 However, molecular analysis of Cameroonian P. falciparum isolates indicated the absence of quintuple mutations (dhfr and dhps) that have been suggested to predict clinical failure after sulfadoxine-pyrimethamine mono-therapy. The limited data on the modulator effect of probenecid on pyrimethamine preclude further clinical studies on this combination until more data on the mechanism of action, pharmacodynamics, pharmacokinetics, and toxicologic studies are available. The apparent discordance between molecular and in vitro assays and in vivo response is probably related to multiple factors, including the synergistic action of pyrimethamine and sulfadoxine, the absence of quintuple mutations and Ile164Leu DHFR substitution in Cameroonian isolates, and acquired immunity. Nevertheless, analysis of molecular markers is an alternative method that characterizes the dhfr alleles of field isolates and may be useful for the long-term monitoring of antifolate-resistant malaria.
Received September 3, 2006. Accepted for publication April 24, 2007.
Acknowledgment: We thank the personnel of Nlongkak Catholic missionary dispensary for their aid in recruiting patients.
Financial support: This study was supported by the French Ministry of Research (Programme PAL+) and the International Scientific Cooperation Project (INCO-DEV) of the European Commission (contract no. ICA4-CT–2001–10078).
* Address correspondence to Leonardo K. Basco, Unité de Recherche Paludologie Afro-Tropicale, Institut de Recherche pour le Développement et Laboratoire de Recherche sur le Paludisme, Organisation de Coordination pour la lutte contre les Endémies en Afrique Centrale, BP 288, Yaoundé, Cameroon. E-mail: lkbasco{at}yahoo.fr 
Authors’ addresses: Rachida Tahar and Leonardo K. Basco, Unité de Recherche Paludologie Afro-Tropicale, Institut de Recherche pour le Développement et Laboratoire de Recherche sur le Paludisme, Organisation de Coordination pour la lutte contre les Endémies en Afrique Centrale, BP 288, Yaoundé, Cameroon, Telephone: 237–2223–22–32 Fax: 237–2223–00–61, E-mails: lkbasco{at}yahoo.fr and rachidatahar{at}yahoo.fr
|
REFERENCES |
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  A. M. McCollum, L. K. Basco, R. Tahar, V. Udhayakumar, and A. A. Escalante Hitchhiking and Selective Sweeps of Plasmodium falciparum Sulfadoxine and Pyrimethamine Resistance Alleles in a Population from Central Africa Antimicrob. Agents Chemother., November 1, 2008; 52(11): 4089 - 4097. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
