CONCISE COMMUNICATION

Transcription

1 624 CONCISE COMMUNICATION Sequence Variations in the Genes Encoding Dihydropteroate Synthase and Dihydrofolate Reductase and Clinical Response to Sulfadoxine- Pyrimethamine in Patients with Acute Uncomplicated Falciparum Malaria Leonardo K. Basco, 1,2 Rachida Tahar, 2,a Annick Keundjian, 3 and Pascal Ringwald 1,a 1 Institut de Recherche pour le Développement and Laboratoire de Recherche sur le Paludisme, Laboratoire Associé Francophone 302, Organisation de Coordination pour la lutte contre les Endémies en Afrique Centrale, Yaoundé, Cameroon; 2 Centre de Génétique Moléculaire, Centre National de la Recherche Scientifique, Gif-sur-Yvette, and 3 Service de Chimie Parasitaire, Institut de Médecine Tropicale du Service de Santé des Armées, Le Pharo, Marseille Armées, France Mutations in dihydropteroate synthase (DHPS) and dihydrofolate reductase (DHFR) are associated with in vitro resistance to sulfadoxine and pyrimethamine, respectively. The response of 75 patients to sulfadoxine-pyrimethamine was determined, and the genes of the corresponding Plasmodium falciparum isolates were sequenced. Of 12 different unmixed allelic combinations, the triple dhfr mutation Asn-108/Arg-59/Ile-51 was observed in all patients responding with early treatment failure. Some, but not all, patients with an adequate clinical response also harbored isolates with the triple dhfr mutation. Higher initial parasitemia and fever distinguished these 2 patient groups. The dhps genotype apparently had no influence on the clinical outcome. The other dhfr alleles with 1 or 2 mutations and the wild-type allele were found in patients with an adequate clinical response. The triple dhfr mutation is one of the genetic determinants associated with in vivo resistance to sulfadoxine-pyrimethamine. Since the rapid spread of chloroquine resistance in Asia and South America in the 1960s and 1970s, sulfadoxine-pyrimethamine has been one of the most important second-line drugs for the treatment of chloroquine-resistant Plasmodium falciparum infections [1]. The extensive use of sulfadoxine-pyrimethamine has led to the selection of resistant strains and declining efficacy to the point that this drug is no longer recommended in some areas endemic for the disease. In Africa, chloroquine resistance emerged much later than in other endemic areas. Chloroquine is still the first-line drug in central and West Africa, and sulfadoxine-pyrimethamine usually is re- Received 29 October 1999; revised 7 April 2000; electronically published 19 July The clinical study was approved by the Cameroonian Ministry of Public Health and the Cameroonian National Ethics Committee. Informed consent was obtained before enrollment from adult patients or guardians of sick children. Financial support: French Ministry of Cooperation and Development; Agence Universitaire pour la Francophonie (AUF, formerly AUPELF- UREF); Zeneca Pharma (La Defense, France). a Present affiliation: Laboratoire d Ecologie Systèmes Vectoriels, Institut Pasteur, Paris, France (R.T.); Cluster of Communicable Diseases, Surveillance and Response, Anti-infective Drug Resistance Surveillance and Containment, World Health Organization, Geneva, Switzerland (P.R.). Reprints or correspondence: Dr. Leonardo Basco, OCEAC/IRD, B.P. 288, Yaoundé, Cameroon (oceac@camnet.cm). The Journal of Infectious Diseases 2000;182: by the Infectious Diseases Society of America. All rights reserved /2000/ $02.00 served for the second-line treatment of chloroquine-resistant infections. In East Africa, however, some countries have adopted the use of sulfadoxine-pyrimethamine for first-line treatment because of the high prevalence of chloroquine resistance [2]. Sulfadoxine-pyrimethamine is a synergistic combination of antifolate drugs that interact with 2 enzymes of the folate biosynthetic pathway. Sulfadoxine competes with p-aminobenzoic acid, the natural substrate of dihydropteroate synthase (DHPS), and inhibits this enzyme. Pyrimethamine binds to malarial dihydrofolate reductase (DHFR) with a high affinity and inhibits the enzymatic activity. The genetic mechanism of in vitro resistance to these antifolates in malaria parasites is due to point mutations in the genes coding for DHPS and DHFR [3, 4]. The key amino acid substitutions in DHPS and DHFR that confer resistance to sulfadoxine and pyrimethamine are Ala- 437-Gly and Ser-108-Asn, respectively. Additional mutations at codons 436, 540, 581, and 613 of DHPS increase the level of sulfadoxine resistance [5]. Likewise, additional mutations at positions 51, 59, and 164 of DHFR are associated with higher levels of pyrimethamine and cycloguanil resistance [6, 7]. In the case of sulfadoxine resistance, an additional mechanism related to folate uptake and utilization may influence resistance levels [8]. In general, previous studies have analyzed the correlation between in vitro resistance to sulfadoxine and dhps genotype

2 JID 2000;182 (August) dhfr and dhps Sequences and In Vivo Response 625 and between in vitro pyrimethamine resistance and dhfr genotype separately in laboratory-adapted reference strains of P. falciparum. The molecular criteria for in vitro antifolate resistance have not been fully validated and correlated with the in vivo response of patients treated with sulfadoxine-pyrimethamine. Furthermore, the allelic combinations of dhfr and dhps that may be associated with therapeutic failure of sulfadoxine-pyrimethamine are not known. In the present study, the in vivo response to sulfadoxine-pyrimethamine was evaluated in Cameroonian patients, and the genes from the corresponding clinical isolates of P. falciparum were fully sequenced with the aim of establishing the correlation between in vivo resistance and the genotypes defined by dhfr and dhps sequences. Materials and Methods The study included 75 patients who met the following inclusion criteria: age 5 years, the presence of fever, monoinfection with P. falciparum, parasite density asexual parasites/ml of blood, absence of signs and symptoms of severe and complicated malaria, and no recent history of self-medication with antimalarial drugs. There were 50 adults (15 59 years old) and 25 children, and the patient group included both sexes. Venous blood samples were obtained before treatment. Parasite DNA was extracted as described elsewhere [9]. The entire DHFR domain (708 bp) of the dhfr-ts gene and part of the DHPS domain (709 bp) of the pppk (hydroxymethyldihydropterin pyrophosphokinase) dhps gene were amplified by polymerase chain reaction (PCR) [9, 10]. The forward primers were phosphorylated at their 5 end by T4 polynucleotide kinase. The 5 -phosphorylated forward primer and the unphosphorylated reverse primer were used to perform PCR. The nucleotide sequences were determined by a single-stranded sequencing protocol, using lambda exonuclease and the dideoxy chain termination reaction [11]. The new standard 14-day test of therapeutic efficacy, which was developed by the World Health Organization (WHO) [12], was used in this study. Sulfadoxine-pyrimethamine was administered under supervision. Adults received a single oral dose of 3 tablets, each containing 500 mg sulfadoxine and 25 mg pyrimethamine, and children received 1 tablet per 20 kg of body weight. The clinical conditions and parasite density were monitored on days 0, 1, 2, 3, 4, 7, and 14 on an outpatient basis. The therapeutic responses were classified as early treatment failure, late treatment failure, or adequate clinical response, according to the criteria set by WHO [12]. Patients who did not respond to sulfadoxine-pyrimethamine treatment were treated with oral halofantrine or quinine. The in vitro response to pyrimethamine was determined by the isotopic microtest [13]. Sulfadoxine and pyrimethamine concentrations were determined simultaneously in plasma samples before treatment and at days 3, 7, and 14 by high-performance liquid chromatography [14]. Results Of the 75 patients successively enrolled and assigned to the sulfadoxine-pyrimethamine treatment group, 9 (12%) were lost to follow-up (table 1). Of the 66 patients who completed the 14-day follow-up, 59 (89%) had an adequate clinical response. One of these patients was afebrile and had negative blood smears on days 2, 3, and 7 and a positive smear on day 14 but was asymptomatic. Genotype analysis based on the amplification of polymorphic genetic markers (merozoite surface antigen 1 and 2 and circumsporozoite protein) by PCR showed that the isolate on day 28 was different from that on day 0, indicating reinfection. The other 58 patients with an adequate clinical response were both afebrile and aparasitemic on or before day 3 and until day 14. Seven patients responded with early treatment failure and required an alternate treatment on or before day 3. None of the patients responded with late treatment failure. The entire DHFR domain and part of the DHPS domain were fully sequenced for all isolates, except for the DHPS domain of 3 isolates obtained from patients who were lost to follow-up. Mutations occurred in codons 51, 59, and 108 of the dhfr-ts gene and codons 436, 437, and 613 of the pppk-dhps gene. The other codons that were reported to have undergone mutation (codons 16 and 164 of the dhfr-ts gene and codons 540 and 581 of the pppk-dhps gene) were invariant and of wildtype codons. An alternate mutation, Thr-108, was not observed in the DHFR domain. Twenty-one isolates had mixed alleles in the DHFR ( n p 11), the DHPS ( n p 13), or both domains. Among the isolates with unmixed alleles in both dhfr and dhps ( n p 48) from patients who completed the 14-day follow-up, 12 different combinations of dhfr and dhps were observed. Isolates that present a single, unmixed allele at all positions allow a correlational analysis between the genotype and phenotype. Of the observed allelic combinations in this study, the consistent feature of the isolates obtained from cases of early treatment failure was the triple dhfr mutation Asn-108/Arg-59/ Ile-51 ( P!.05; Fisher s exact test). Six of 7 isolates from patients with early treatment failure had the triple dhfr mutation, whereas 1 isolate had mixed alleles, including the triple dhfr mutation. Three different dhps alleles were associated with the triple dhfr mutation in these isolates: mutant allele Ala-436/ Ala-437/Lys-540/Ala-581/Ala-613 ( n p 2), mutant allele Ser- 436/Gly-437/Lys-540/Ala-581/Ala-613 ( n p 3), and mixed allele Ala- and Ser-436/Ala-437/Lys-540/Ala-581/Ala-613 ( n p 1). In addition, the dhps allele Cys-436/Ala-437/Lys-540/Ala-581/Ala- 613 associated with the triple Asn-108/Arg-59/Ile-51 dhfr mutation was detected in 1 isolate. In this patient, the pretreatment isolate displayed mixed dhfr alleles at positions 51, 59, and 108 and a pure dhps allele (Cys-436/Ala-437/Lys-540/Ala-581/Ala- 613). The day-3 isolate had an unmixed dhfr allele with the triple mutation and an identical dhps allele as the day-0 isolate. The 2 allelic combinations observed in early treatment failure (triple dhfr mutation and either Ala-436/Ala-437/Lys-540/Ala- 581/Ala-613 or Ser-436/Gly-437/Lys-540/Ala-581/Ala-613 dhps alleles) were also observed in 8 (1 from a patient with a positive blood smear on day 14 and without fever) and 7 isolates, re-

3 626 Basco et al. JID 2000;182 (August) Table 1. dhfr and dhps alleles and in vivo response to sulfadoxine-pyrimethamine in Cameroonian patients with acute uncomplicated falciparum malaria. In vivo response No. of patients DHFR amino acid residues DHPS amino acid residues DHFR Genbank no. DHPS Genbank no. ACR 2 A N C S I S A K A A AF AF ACR 10 A N C S I A A K A A AF AF ACR 4 A N C S I S G K A A AF AF ACR 4 A N C N I A A K A A AF AF ACR 2 A N R N I S A K A A AF AF ACR 3 A N R N I A A K A A AF AF ACR 2 A N R N I S G K A A AF AF ACR a 8 A I R N I A A K A A AF AF ACR 7 A I R N I S G K A A AF AF ACR 1 A I R N I A G K A A AF AF ACR 1 A N C S I S/A G K A A AF AF ACR b 1 A N C S I S/F A K A A/S AF AF ACR 1 A I R N I S/A A K A A AF AF ACR 1 A I R N I S/A G K A A AF AF ACR 1 A I R N I A A/G K A A AF AF ACR 2 A I R N I S/A A/G K A A AF AF ACR 1 A N C/R S/N I A A K A A AF AF ACR 1 A N C/R S/N I S/A A/G K A A AF AF ACR 1 A N C S/N I A A K A A AF AF ACR 1 A N C S/N I S/A A K A A AF AF ACR 1 A N/I C/R S/N I S A K A A AF AF ACR 2 A N/I C/R S/N I A A K A A AF AF ACR 1 A N/I C/R S/N I S/A A K A A AF AF ACR 1 A I R S/N I S G K A A AF AF ETF 2 A I R N I A A K A A AF AF ETF 3 A I R N I S G K A A AF AF ETF 1 A I R N I S/A A K A A AF AF ETF 1 A N/I C/R S/N I C A K A A AF AF ND 1 A N C S I A A K A A AF AF ND 1 A N C S I F A K A S AF AF ND 1 A N C S I S/A G K A A AF AF ND 1 A I R N I S G K A A AF AF ND 1 A I R N I S/A A/G K A A AF AF ND 1 A N/I C/R S/N I A A K A A AF AF ND 3 A I R N I AF NOTE. In vivo response was determined by in vivo test of therapeutic efficacy developed by the World Health Organization. Nine patients were lost to follow-up. Boldface indicates mutant codons. ACR, adequate clinical response; ETF, early treatment failure; ND, not determined. a One of the patients had asymptomatic parasitemia on day 14. b The dhps sequence of this isolate had a 3 amino acid deletion (ISA, positions ) and a 3 amino acid insertion (LLK, positions ). spectively, that were derived from patients responding with an adequate clinical response. The other unmixed combinations involving the wild-type or single and double dhfr mutant alleles associated with Ser- or Ala-436/Ala- or Gly-437 alternative dhps mutations (n p 27) were all observed in isolates from patients responding to sulfadoxine-pyrimethamine treatment. The clinical and laboratory parameters of all patients infected with isolates carrying the triple dhfr mutation (27 with pure triple mutant allele and 5 with mixed alleles with all 3 mutations) were compared in relation to the clinical outcome (table 2). The pretreatment body temperature and parasitemia were higher in patients responding with early treatment failure ( P!.05; t test). The neutrophil count and total bilirubin were also significantly elevated ( P!.05; t test) in patients who did not respond to sulfadoxine-pyrimethamine treatment. The other clinical and laboratory measures and plasma drug concentrations were not significantly different between the 2 groups ( P 1.05; t test). In vitro assays for pyrimethamine yielded interpretable results for 56 of 75 isolates (table 3). All 17 isolates with the wildtype dhfr allele were characterized by low IC 50 values (!28.5 nm; mean 1.10 nm). By contrast, all isolates with mutant dhfr alleles, with the exception of those presenting mixed alleles, had IC 50 values 170 nm. Although there were some overlapping values, the mean IC 50 values among mutant isolates increased 3 4-fold with the number of point mutations. The mean pyrimethamine IC 50 was 106 nm with single mutation, 308 nm with double mutation, and 1360 nm with triple mutation. Discussion The triple dhfr mutation Asn-108/Arg-59/Ile-51 may be associated with either an adequate clinical response or early treat-

4 JID 2000;182 (August) dhfr and dhps Sequences and In Vivo Response 627 Table 2. Pretreatment clinical and laboratory characteristics and plasma drug concentrations in patients infected with Plasmodium falciparum isolates carrying the triple dhfr mutation. Parameter Adequate clinical response (n p 25) Early treatment failure (n p 7) Sex ratio (M/F) 16/9 3/4 No. of children ( 5 years old)/adult 5/20 4/3 Age, years (10 59) (11 37) Weight, kg (30 88) (41 70) Body temperature, C ( ) ( ) Parasitemia, asexual parasites/ml blood 31,000 ( ,000) 113,000 (21, ,000) Hemoglobin, g/dl ( ) ( ) White blood cell count, 10 9 /L ( ,100) ,460 ( ,600) Neutrophil count, 10 9 /L ( ) ( ,100) Lymphocyte count, 10 9 /L ( ) ( ) Platelet count, 10 9 /L (59 227) (75 169) ASAT, IU/L (14 61) (19 42) ALAT, IU/L (8 69) (17 49) Total bilirubin, mm ( ) ( ) Blood urea nitrogen, mm ( ) ( ) Blood creatinine, mm (42 124) (44 142) Pyrimethamine, ng/ml Day ( ) ( ) Day ( ) ( ) Day (0 34.3) (0 51.3) Sulfadoxine, mg/ml Day ( ) ( ) Day ( ) ( ) Day ( ) ( ) NOTE. Values are mean SD (range). Parasitemia is expressed as the geometric mean. The following parameters differed significantly ( P!.05; t test) between the 2 groups: pretreatment body temperature, initial parasitemia, absolute neutrophil and lymphocyte counts, and total bilirubin. ASAT, aspartate aminotransferase; ALAT, alanine aminotransferase. ment failure in Cameroonian children 5 years old and adults residing in Yaoundé. Conversely, when a patient responds with early treatment failure to sulfadoxine-pyrimethamine therapy, it can be predicted that the isolate includes at least 1 parasite population with the triple dhfr mutant allele; however, when a patient responds with an adequate clinical response, no one particular dhfr genotype can be predicted, and various combinations of dhfr and dhps genotypes can be expected. If the triple dhfr mutation is the main genetic determinant of therapeutic failure with sulfadoxine-pyrimethamine, the next question that arises concerns additional factors that determine whether a given patient infected with the triple dhfr mutant isolate can eliminate parasitemia. Among pretreatment clinical features that were analyzed in patients, fever and elevated parasitemia seemed to be predictive of a treatment failure. Although there is no absolute correspondence between high parasitemia in the peripheral circulation and severity of the disease, elevated parasitemia associated with high fever may generally indicate a higher degree of risk for severe malaria. Another parameter that cannot presently be quantitated but that can be hypothesized to play a determinant role is the level of acquired immunity. Patients may have a different immune capacity (independent of the drug-resistant phenotype of the parasites) to eliminate parasites, and the level of acquired immunity is related to both age and exposure to malaria infections. The pharmacodynamic and pharmacokinetic variations may also account for differential clinical responses to the treatment, regardless of the genotypic features of the infecting para- Table 3. Comparison of dhfr genotypes and in vitro pyrimethamine sensitivity levels for 56 Plasmodium falciparum isolates. Pyrimethamine sensitivity Sensitive Moderately resistant Highly resistant dhfr Genotype n Mean IC 50 n Mean IC 50 n Mean IC 50 Wild-type ( ) 0 0 Single mutation ( ) 0 Double mutation ( ) 0 Triple mutation ( ) ( ,600) Mixed ( ) ( ) 0 NOTE. Single mutation, Asn-108; double mutation, Asn-108/Arg-59; triple mutation, Asn-108/Arg- 59/Ile-51. Values are geometric mean (range). The pyrimethamine IC 50 values were arbitrarily classified as sensitive (!100 nm), moderately resistant ( nm), or highly resistant (12000 nm).

5 628 Basco et al. JID 2000;182 (August) sites. In our study, however, patients infected with triple dhfr mutant isolates had comparable plasma concentrations of pyrimethamine and sulfadoxine. In conclusion, our results suggest that the triple dhfr mutation Asn-108/Arg-59/Ile-51 might be a genetic marker for in vivo resistance to sulfadoxine-pyrimethamine in some patients in central Africa and that the genotypic pattern of dhps may play a secondary role in determining treatment failure. However, the triple mutation is clearly not the sole determinant that accounts for therapeutic failure. Although conclusive evidence cannot be derived from the present study, other parasite- and host-related factors need to be examined in future studies. Acknowledgments We thank Sisters Solange Menard and Marie-Solange Oko and their nursing and laboratory staff at the Nlongkak Catholic missionary dispensary for precious aid in recruiting patients. References 1. Peters W. Chemotherapy and drug resistance in malaria. 2d ed. London: Academic Press, Bloland PB, Lackritz EM, Kazembe PN, Were JB, Steketee R, Campbell CC. Beyond chloroquine: implications of drug resistance for evaluating malaria therapy efficacy and treatment policy in Africa. J Infect Dis 1993;167: Hyde JE. The dihydrofolate reductase thymidylate synthetase gene in the drug resistance of malaria parasites. Pharmacol Ther 1990;48: Brooks DR, Wang P, Read M, Watkins WM, Sims PFG, Hyde JE. Sequence variation of the hydroxymethyldihydropterin pyrophosphokinase: dihydropteroate synthase gene in lines of the human malaria parasite, Plasmodium falciparum, with differing resistance to sulfadoxine. Eur J Biochem 1994;2: Triglia T, Menting JGT, Wilson C, Cowman AF. Mutations in dihydropteroate synthase are responsible for sulfone and sulfonamide resistance in Plasmodium falciparum. Proc Natl Acad Sci USA 1997;94: Foote SJ, Galatas D, Cowman AF. Amino acids in the dihydrofolate reductase thymidylate synthase gene of Plasmodium falciparum involved in cycloguanil resistance differ from those involved in pyrimethamine resistance. Proc Natl Acad Sci USA 1990;87: Peterson DS, Milhous WK, Wellems TE. Molecular basis of differential resistance to cycloguanil and pyrimethamine in Plasmodium falciparum malaria. Proc Natl Acad Sci USA 1990;87: Wang P, Brobey RKB, Horii T, Sims PFG, Hyde JE. Utilization of exogenous folate in the human malaria parasite Plasmodium falciparum and its critical role in antifolate drug synergy. Mol Microbiol 1999;32: Basco LK, Ringwald P. Molecular epidemiology of malaria in Yaoundé, Cameroon. VI. Sequence variations in the Plasmodium falciparum dihydrofolate reductase thymidylate synthase gene and in vitro resistance to pyrimethamine and cycloguanil. Am J Trop Med Hyg 2000;62: Basco LK, Ringwald P. Molecular epidemiology of malaria in Yaoundé, Cameroon. II. Baseline frequency of point mutations in the dihydropteroate synthase gene of Plasmodium falciparum. Am J Trop Med Hyg 1998;58: Ausubel FM, Brent R, Kingston RE, et al. Short protocols in molecular biology. 3d ed. New York: John Wiley & Sons, World Health Organization (WHO). Assessment of therapeutic efficacy for uncomplicated falciparum malaria in areas with intense transmission. Geneva: WHO, 1996; WHO/MAL/ Desjardins RE, Canfield CJ, Haynes JD, Chulay JD. Quantitative assessment of antimalarial activity in vitro by a semiautomated microdilution technique. Antimicrob Agents Chemother 1979;16: Edstein M. Simultaneous measurement of sulphadoxine, N 4 acetylsulphadoxine and pyrimethamine in human plasma. J Chromatogr 1984;305: