Mem Inst Oswaldo Cruz, Rio de Janeiro, VOLUME 114 | JAN 2019
Lack of quadruple and quintuple mutant alleles associated with sulfadoxine-pyrimethamine resistance in Plasmodium vivax isolates from Brazilian endemic areas
1Fundação Oswaldo Cruz-Fiocruz, Instituto Oswaldo Cruz, Laboratório de Pesquisa em Malária, Rio de Janeiro, RJ, Brasil
2Fundação Oswaldo Cruz-Fiocruz, Centro de Pesquisa, Diagnóstico e Treinamento em Malária, Rio de Janeiro, RJ, Brasil
3Ministério da Saúde, Secretaria de Vigilância em Saúde, Programa Nacional de Prevenção e Controle da Malária, Brasília, DF, Brasil
4Fundação Oswaldo Cruz-Fiocruz, Instituto Nacional de Infectologia Evandro Chagas, Laboratório de Doenças Febris Agudas, Rio de Janeiro, RJ, Brasil
5Institut Pasteur, Malaria Genetic and Resistance Group, Biology of Host-Parasite Interactions Unit, Paris, France
BACKGROUND AND OBJECTIVE Brazil is responsible for a large number of Plasmodium vivax cases in America. Given the emergence of P. vivax parasites resistant to chloroquine and the effectiveness of antifolates in vivax malaria treatment together with a correlation between mutations in P. vivax dhfr and dhps genes and SP treatment failure, the point mutations in these genes were investigated.
METHODS Blood samples from 54 patients experiencing vivax malaria symptomatic episodes in the Amazonian Region were investigated. Genomic DNA was extracted using a DNA extraction kit (QIAGENTM). Nested polymerase chain reaction (PCR) amplification was carried out followed by Sanger sequencing to detect single nucleotide polymorphisms (SNPs).
FINDINGS All tested isolates showed non-synonymous mutations in pvdhfr gene: 117N (54/54, 100%) and 58R (25/54, 46%). Double mutant allele 58R/117N (FRTNI, 28%) was the most frequent followed by triple mutant alleles (58R/117N/173L, FRTNL, 11%; 58R/61M/117N, FRMNI, 5% 117N/173L, FSTNL, 4%) and quadruple mutant allele (58R/61M/117N/173L, FRMNL, 2%). A single mutation was observed at codon C383G in pvdhps gene (SGKAV, 48%).
CONCLUSION No evidence of molecular signatures associated with P. vivax resistance to SP was observed in the Brazilian samples.
Plasmodium vivax is the most geographically widespread human malaria parasite. It is prevalent mainly outside Africa including Asia, South and Central America, and the Middle East. In the Americas, the burden of vivax malaria mostly affects Venezuela and Brazil. In Brazil, malaria transmission occurs almost entirely (> 99% of the registered cases) within the northern Brazilian Amazon Region where both P. falciparum and P. vivax infections co-exist. In this area, P. vivax is the predominant species, responsible for 89% of 194,409 malaria cases reported in 2017.(1) Nowadays, falciparum malaria is treated with a 3-day fixed Artesunate+Mefloquine combination, according to Brazilian National Malaria Program guidelines, and a radical cure for P. vivax malaria is achieved with 25 mg/kg of CQ base for three days (maximum adult dose, 1.5 g for three days), combined with a short hypnozoitocidal regimen of 0.5 mg/kg/day of primaquine (PQ) base (maximum daily dose, 30 mg/day) for seven days in patients that weighed below 70 kg. As subtherapeutic PQ doses may lead to relapse in overweight patients, weight-adjusted PQ doses are now recommended in Brazil for patients over 70 kg.
P. falciparum resistance to chloroquine (CQ) observed in the 1980s greatly contributed to the emergence of falciparum malaria outbreaks across Amazon.(2) P. vivax resistance to CQ occurred later in 1989 in Papua New Guinea(3) and CQ monotherapy was ineffective. Following this seminal observation, numerous cases of CQ resistance were reported in Southeast Asia(4) and South America,(5, 6) thus complicating the current international efforts for malaria control and elimination, and signalling the need for alternative drugs for vivax malaria treatment.
Antifolates, most notably sulfadoxine-pyrimethamine (SP), have been used as anti-malaria for P. falciparum treatment throughout the world because this combination is inexpensive, relatively safe, and requires only a single dose course treatment. SP had been available in Brazil since 1960s to treat CQ-resistant falciparum malaria but SP-resistant P. falciparum isolates appeared since 1990; SP is not used for malaria therapy in Brazil. Although resistant to antifolates, P. falciparum treatment has been well documented in many parts of the world, and P. vivax chemoresistance to SP is scarcely studied.
Sulfadoxine and pyrimethamine are competitive inhibitors of dihydropteroate synthase (dhps) and dihydrofolate reductase (dhfr), the two major proteins involved in folate biosynthesis pathway(7). Polymorphisms in these two genes are the major factors associated with SP resistance.
Data on pvdhfr and pvdhps genotypes are available for many Southeast Asian countries. Such reports remain limited for some P. vivax endemic areas, notably South America. In Brazil, only one study characterising polymorphisms in pvdhfr gene was documented8) and there is no report on the frequency of single nucleotide polymorphism (SNP) in dhps gene in P. vivax clinical isolates from Brazilian endemic areas.
Given the emergence of P. vivax CQ resistant parasites and the effectiveness of antifolates in malaria vivax treatment together with a strong correlation between mutations in P. vivax dhfr and dhps genes and SP treatment failure,(9) the present paper reports an investigation on the pattern of point mutations in pvdhfr and pvdhps genes in Brazilian isolates.
MATERIALS AND METHODS
Parasites isolates and DNA extraction - Blood samples from Amazon Region (Acre, Amapá, Amazonas, Rondônia and Pará) were collected from 54 patients presenting with vivax malaria from 2010 to 2016 at the Laboratório de Doenças Febris Agudas, INI-IPEC, Fiocruz, the Reference Clinical Laboratory for Malaria in the Extra-Amazon to the Brazilian Ministry of Health. All the clinical isolates were diagnosed as single P. vivax infections by light microscopic examination of Giemsa’s solution-stained blood smears and by P. vivax cysteine-proteinase target gene polymerase chain reaction (PCR).(10) The parasitaemia ranged from 960 to 19160 parasites/µL. All malaria patients presented with clinical signs and/or symptoms of uncomplicated malaria, such as fever, headache, and chills, and the baseline characteristics were similar. No significant difference in parasitaemia was observed among the studied Brazilian localities and all the Brazilian endemic states were hypoendemic malaria areas.
Genomic DNA was extracted using a commercially available DNA extraction kit (QIAGENTM, Frankfurt, Germany), following the manufacturer’s instructions. This study was performed according to the protocols previously approved by the Ethical Research Committees of Fiocruz (32839013.6.00005248). Patients were treated with CQ plus PQ, according to the Brazilian Ministry of Health recommendation for uncomplicated vivax malaria treatment and were followed up to 42 days. No treatment failure was detected during this period.
Nested PCR and electrophoresis - Nested PCR amplification of pvdhfr and pvdhps were carried out as described previously.(11) Ten point mutations were investigated: F57L/I, S58R, T61M, S117T/N and I173F/L for pvdhfr,and S382A, C383G, K512M/T/E, A553G and V585G for pvdhps. PCR products were analysed by ethidium bromide-stained agarose-gel (2%) electrophoresis.
DNA sequencing and SNPs detection - The 632 bp and 767 bp fragments generated by amplification of pvdhfr and pvdhps, respectively, were extracted and purified from gel using the Wizard® SV Gel and PCR Clean-Up System (Promega, Madison, WI, USA) commercial kits. Briefly, the amplified fragments were sequenced using BigDye Terminator cycle sequencing ready reaction version 3.1 and ABI Prism DNA analyser 3730 (Applied Biosystems) at the Genomic Platform/PDTIS/Fiocruz. The direct DNA sequencing from PCR products were compared with the reference Sal I sequence of pvdhfr (GenBank X98123) and pvdhps (GenBank AY186730.1). Forward and reverse sequences were analysed using the free software, Bioedit Sequence Alignment Editor version 7.2.5. PCRs and DNA sequencing were randomly repeated to check possible sequence errors introduced during these stages.
All the 54 isolates sequenced for pvdhfr gene showed non-synonymous mutations: 117N (54/54; 100%) and 58R (25/54; 46%) mutant alleles were more frequent, while 173L (9/54; 17%) and 61M (4/54; 7%) were detected at lower frequencies. Mutation at position 57L was not found (Table I). The most common single mutant allele was 117N (27/54; 50%). This single mutant was more frequent in Acre (10/15; 66%), Amazonas (11/23; 52%) and Pará states (4/8; 50%), compared to Rondônia state (1/7; 14%), where double 58R+117N mutant was dominant (Table II). Independent of the year collection, Amazonas state showed the highest number of pvdhfr gene mutations (23/54; 42,5%), followed by Acre (15/54; 27,7%, Para (8/54; 15%) and Rondônia (7/54, 13%) (Tables III-VI). Apparently in 2011, Acre presented more pvdhfr gene mutations (7/15; 47%) than Amazonas (2/23; 8,6%) (Tables III-IV), but this difference could be related to the smaller number of Amazonas samples collected in 2011, because when percentages are compared instead of figures, 100% of Amazonas (2/2) and Acre samples (7/7) presented mutations in 2011.
The double 58R/117N allele (FRTNI, 28%) was the most common allele, contrasting with the frequencies of other dhfr double, triple, or quadruple mutant alleles, with lower frequencies: 58R/117N/173L (FRTNL, 11%), 58R/61M/117N (FRMNI, 5%), 117N/173L (FSTNL, 4%), and 58R/61M/117N/173L (FRMNL, 2%). In all localities, wild-type pvdhfr (FSTSI) was not observed (Table VII). The 58R/117N double mutant allele was detected in Acre (2/15; 13%), Rondônia (5/7; 71%), and Amazonas (8/23; 35%) while the 117N+173L only in Pará and Rondônia. The triple mutant allele 58R/117N/173L was found in all localities, except Rondônia, and the quadruple mutant 58R/61M/117N/173L was observed onl)in one isolate collected from Amazonas state (1/23; 4%) (Table VIII). The frequencies of double, triple, or quadruple mutants were not related to the year of collection (Tables III-VI).
Concerning pvdhps gene in 26 out of 54 (48%) isolates only a single mutation at codon C383G was detected. No other mutations, including 382A, 512M, 553G, and 585C, were found. The wild-type SCKAV (52) and single haplotype SGKAV (48%) were observed at similar frequencies. The single mutant 383G was observed in isolates from Amazonas (13/23, 56%), Acre (8/15, 53%) and Pará (5/8, 62%) but not in isolates from Rondônia state (0/7) (Table II). Once again frequencies of pvdhps gene mutations were not related to the year of collection (Tables III-VI).
Combining pvdhfr and pvdhps alleles, only one haplotype (FRTNI for pvdhfr and SGKAV for pvdhps) was seen in three of the four study sites with a higher frequency in Amazonas state (where one pvdhfr qu)ruple mutant was detected) (Table IX). No pvdhfr or pvdhps quadruple or quintuple mutant haplotype, which might result in poor clinical response against antifolate drugs, was detected in any of the Brazilian localities investigated.