Polymorphism in drug resistance genes dihydrofolate reductase and dihydropteroate synthase in Plasmodium falciparum in some states of India

Background Sulfadoxine-pyrimethamine (SP) combination drug is currently being used in India for the treatment of Plasmodium falciparum as partner drug in artemisinin-based combination therapy (ACT). Resistance to sulfadoxine and pyrimethamine in P. falciparum is linked with mutations in dihydropteroate synthase (pfdhps) and dihydrofolate reductase (pfdhfr) genes respectively. This study was undertaken to estimate the prevalence of such mutations in pfdhfr and pfdhps genes in four states of India. Methods Plasmodium falciparum isolates were collected from two states of India with high malaria incidence i.e., Jharkhand and Odisha and two states with low malaria incidence i.e., Andhra Pradesh and Uttar Pradesh between years 2006 to 2012. Part of sulfadoxine-pyrimethamine (SP) drug resistance genes, pfdhfr and pfdhps were PCR-amplified, sequenced and analyzed. Results A total of 217 confirmed P. falciparum isolates were sequenced for both Pfdhfr and pfdhps gene. Two pfdhfr mutations 59R and 108N were most common mutations prevalent in all localities in 77 % of isolates. Additionally, I164L was found in Odisha and Jharkhand only (4/70 and 8/84, respectively). Another mutation 51I was found in Odisha only (3/70). The pfdhps mutations 436A, 437G, 540E and 581G were found in Jharkhand and Odisha only in 13, 26, 14 and 13 % isolates respectively, and was absent in Uttar Pradesh and Andhra Pradesh. Combined together for pfdhps and pfdhfr locus, triple, quadruple, quintuple and sextuple mutations were present in Jharkhand and Odisha while absent in Uttar Pradesh and Andhra Pradesh. Conclusion While only double mutants of pfdhfr was present in low transmission area (Uttar Pradesh and Andhra Pradesh) with total absence of pfdhps mutants, up to sextuple mutations were present in high transmission areas (Odisha and Jharkhand) for both the genes combined. Presence of multiple mutations in pfdhfr and pfdhps genes linked to SP resistance in high transmission area may lead to fixation of multiple mutations in presence of high drug pressure and high recombination rate. Electronic supplementary material The online version of this article (doi:10.1186/s13071-015-1080-2) contains supplementary material, which is available to authorized users.


Background
Malaria is one of the major health problems in tropical and subtropical countries. One of the greatest challenges to malaria treatment is the development and spread of resistance in parasites especially in Plasmodium falciparum which threaten the usable lifespan of even artemisininbased combination therapies, affecting both the artemisinin component and the partner medicine [1]. India has evidenced resistant parasite especially P. falciparum against all available conventional antimalarials like chloroquine (CQ) and sulfadoxine-pyrimethamine (SP) [2]. A decade long use of artemisinin-based combination therapy (ACT) had been proved a hallmark anti-malarial therapy for all the malaria endemic countries [3]. The reports of chloroquine resistance in P. falciparum in early 1980s lead to introduction of SP as a second line antimalarial drug in CQ-resistant areas of India [4]. Sulfadoxine and pyrimethamine acts as a synergistic combination and was used as long acting partner antimalarial drug in ACT in South Asia, Middle East and South America [3]. Since 2005, Indian antimalarial drug policy has introduced artesunate with SP as ACT in place of SP in high malaria endemic areas, and later in 2010, this treatment became the recommended first line treatment throughout India [5,6]. Further, since 2013, prevalence of resistant genotype of falciparum against this partner drug SP, led to introduction of artemether-lumefantrine as antimalarial therapy for northeastern part of India [7,8].
The synergistic combination of the sulfadoxine and pyrimethamine inhibits dihydropteroate synthase (dhps) and dihydrofolate reductase (dhfr) enzymes respectively in the folate-pathway of parasite [9]. The development of resistance against SP emerges with a single point mutation in the parasite dhfr and dhps gene, which further augments with stepwise addition of mutations [10][11][12]. Resistance to pyrimethamine is primarily conferred by a point mutation at codon 108 and augmented by mutations at codon 16, 51, 59 and 164 of Plasmodium falciparum dhfr (pfdhfr) gene [10,12,13]. Similarly, point mutation at codon 436 or 437 in Plasmodium falciparum dhps (pfdhps) gene may initiate the resistance and followed by mutations at codon 540, 581 and 613, which are considered for augmentation of sulfadoxine resistance [11,13,14].
A single mutation in pfdhfr or pfdhps gene is not enough to cause treatment failure and multiple mutation combinations in these two genes were associated with failure of SP as anti-malarial therapy [15,16]. Various parts of India reported single, double, triple and quadruple mutant pfdhfr gene [17][18][19][20][21][22]. However, double mutation at codon 59 and 108 in pfdhfr gene was predominant throughout India [17][18][19][20][21][22]. Triple mutant pfdhfr gene indicating high level of antifolate resistance was observed in India from northeast states, Car Nicobar island and Odisha [7,15,20,22,23]. Highly resistant quadruple mutant allele was observed in high and low frequencies from Car Nicobar island and northeastern parts of India respectively [19,20,22,24,25]. Wild-type allele in pfdhps gene was predominant in all geographic regions of India except Andaman and Nicobar island, where lower frequency of mutations in pfdhps gene was observed in comparison to pfdhfr gene, which supports that mutations first emerged in pfdhfr and then occurs in pfdhps gene [23]. Single mutation at codon 437 was observed in low frequency from Assam, Odisha, Madhya Pradesh and Uttar Pradesh [21,23,26]. Further, double and triple mutation including mutation at codon 437 was also observed in low frequencies from Madhya Pradesh, northeast and Odisha [20,26,27]. However, recent studies from northeastern part of India showed increased number of key mutation at codon 437 included in triple and quadruple mutations in pfdhps gene [7,20,22].
Development of resistance against ACT is currently a major threat and P. falciparum bearing resistance against its partner drug (here, SP) may lead to ACT failure [3]. The reports of widespread resistance against SP generate concern about long-term effectiveness of ACT in India [5,20]. A recent study reported significant reduction in efficacy of SP treatment from northeastern areas of India, which is considered as a gateway for invasion of drug resistant parasite from Southeast Asia to India [7,20,22]. Thus, routine molecular surveillance of SP resistance markers is essential in malaria endemic regions, which will help in formulating an effective malaria treatment strategy. Here, we attempted to determine the changes in the frequencies of dhfr and dhps mutations in P. falciparum isolates from four states of India (Jharkhand, Odisha, Andhra Pradesh and Uttar Pradesh) to assess the level of SP resistance.

Study population and blood sample collection
Finger prick blood-spots (n = 217) were collected on Whatman 3MM filter paper from all microscopicallyconfirmed P. falciparum positive patients. This study has been approved by the Institutional Ethics Committee (IEC), and the Scientific Advisory Committee (SAC) of National Institute of Malaria Research. All isolates were collected between years 2006-2012. The study included patients with symptoms of uncomplicated malaria, visiting Primary Health Centre (PHC) situated in the six districts of the four different states of India, i.e., Jharkhand, Odisha, Andhra Pradesh and Uttar Pradesh which are described below. Artemesinin-based combination therapy was implemented as the first line of antimalarials drug for treatment of P. falciparum in some districts of Andhra Pradesh, Jharkhand and Orissa since 2008 and in Uttar Pradesh since 2010. Samples were collected from following sites: . Ghaziabad is an industrial area, located in between three river named Hindon, Ganga and Yamuna. Malaria endemicity is low with seasonal transmission and P. vivax is the predominant malaria species in this region. API of Ghaziabad is ≥2 [29]. Sample (n = 31) were collected from this site during years 2011 and 2012; during that time only 3.25 % (n = 1857) and 1.56 % (n = 740) of total malaria burden was attributed to P. falciparum in same years respectively [30].

DNA isolation and molecular diagnosis
Genomic DNA was extracted from dried filter paper blood spots using QIAmp Blood mini kit (Qiagen, Krefeld, Germany) as per the manufacture's instruction. A PCR diagnosis was performed to confirm the presence of P.
falciparum infection and rule out any mixed species infection as described earlier [31].
SNP's genotyping in pfdhfr and pfdhps genes PCR amplification of pfdhfr gene A 720 base pair fragment of pfdhfr gene was amplified as described earlier [18] and nested PCR was performed to amplify 648-bp fragment covering various single nucleotide polymorphism (SNP) A16V, N51I, C59R, S108N and I164L correlated with pyrimethamine resistance. Primary PCR (25 μL) reaction contained 4 μl DNA template, 200 μM of dNTPs, 1x PCR buffer, 0.30 μM of primers AMP-1 F and AMP-2R, and 2.5 U Taq polymerase (Sigma, India). The cycling parameters used were as follows: an initial denaturation at 94°C for 3 min, denaturation at 94°C for 30 s, annealing at 45°C for 45 s, and extension at 72°C for 45 s, for 45 cycles followed by an extension step at 72°C for 5 min. Nested PCR was performed using 2.5 μl templates from first round PCR product, 0.30 μM of primers M1 and M5, 200 μM of each dNTP, 1x PCR buffer, and 1 U of Taq polymerase in a 25 μl reaction. The PCR was carried at 94°C for 3 min, followed by 35 cycles at 94°C for 1 min, 45°C for 1 min, 72°C for 1 min and finally 72°C for 10 min. The PCR amplicon was visualized in 2 % Agarose gel. The list of primers used for PCRamplification is provided in Table 1.

PCR amplification of pfdhps gene
A nested PCR assay was performed to amplify 728-bp fragment of the pfdhps gene covering SNP's S436A, A437G, K540E, A581G and A613S known to be associated with sulfadoxine resistance as described earlier [18]. Primary PCR reaction of 25 μL was prepared consisting 4 μl DNA template, 200 μM of each dNTP, 1x PCR buffer, 0.30 μM of primers M3717F and 186R and 2.5 U Taq polymerase (Sigma, India). The cycling parameters used were as follows: an initial denaturation at 94°C for 5 min, denaturation at 94°C for 30 s, annealing at 55°C for 45 s, and extension at 72°C for 90 s, for 45 cycles followed by an extension step at 72°C for 10 min. Nested PCR  Table 1.

DNA sequencing and analysis
All successful nested PCR amplicons were purified using MinElute PCR Purification Kit (Qiagen, Krefeld, Germany) and subjected to DNA sequencing using Big-Dye Terminator Kit version 3.1 (Applied Biosystems, Foster, USA). Sequencing was performed on both strand of DNA to confirm SNP's. The chromatogram was manually edited in Finch TV and mixed bases were carefully scored. Sequences obtained were aligned in software MEGA version 5 [32] using ClustalW implemented in the program with the wild type sequence obtained from GenBank with Accession number J03028.1 and PFU07706 for pfdhfr and pfdhps respectively. DNA sequences obtained were submitted to GenBank (accession numbers KP30040 -KP300256 for pfdhfr and KP300257 -KP300473 for pfdhps gene).
The genetic diversity parameters such as haplotype diversity and two measures of nucleotide diversity; θ w and π were estimated in software DnaSP version 5.10.01. The estimation of θ w and π is based on the number of segregating sites and mean number of pairwise nucleotide differences respectively. To test the neutrality in molecular evolution of pfdhfr and pfdhps gene, Tajima's D was calculated based on the normalized discrepancy between θ w and π. Other measures of neutrality such as Fu and Li's D* and Fu and Li's F* were also evaluated.

Sulfadoxine-Pyrimethamine resistance genotypes
Mutations at both genes were concatenated to form combined mutant genotypes, which in turn provide information about various levels of clinical resistance to SP treatment. In an earlier study [18], the combined mutant genotypes were categorized for following types of clinical resistance; category "S/RI" with single or double mutation in combined genotype infers early emergence of SP resistance; category "RI" with triple mutation in combined genotype suggests low level of SP-resistance; category "RI/RII" with triple mutation in pfdhfr gene suggests initiation of high level SP-resistance; category "RII/RIII" with quintuple mutations in combined genotype suggests higher level of SP-resistance; category "RIII" with sextuple mutations in combined genotype suggests total failure of SP treatment.

Results
Mutation analysis of pfdhfr and pfdhps genes

Two-locus mutation status of pfdhfr and pfdhps genes
A total of 13 different pfdhfr-pfdhps two locus genotypes were observed among 217 isolates which are presented in Table 4. Wild-type two locus genotypes were observed in 17.51 % (n = 38) isolates. Mutant two locus genotypes were observed in 82.48 % (n = 179) isolates. Out of that, 62.6 % (n = 136) isolates showed mutant pfdhfr in the two-locus combination. The majority of isolates (115 of 136) had double mutant pfdhfr (GEN3) while single and triple mutant pfdhfr were found in 12 (GEN2) and 9 (GEN7 and GEN8 ) isolates respectively. Mutant genotype for both genes in the two locus combination was observed in 19.81 % (n = 43) isolates. The two locus combined mutant genotypes were categorized into various levels of clinical resistance as described earlier [18], which was based on number of mutations observed in the combination ( Table 4). The category "S/RI" representing emergence of SP resistance was observed as genotype GEN2 and GEN3 in 58.52 % (n = 127) isolates. The category "RI" suggested low level of resistance against SP treatment was observed as genotype GEN4-6 in 2.30 % (n = 5) isolates. The category "RI/RII" suggested for high level of resistance against SP treatment was observed as genotype GEN7 in 1.38 % (n = 3) isolates. The category "RII" with quadruple mutation in two locus combination suggested for higher level of resistance against SP treatment was observed as genotype GEN8-10 in 11.52 % (n = 25) isolates. The category "RII/RIII" with quintuple mutation in two-locus combination also suggested for higher level of resistance against SP treatment was observed as genotype GEN11 in 5.53 % (n = 12) isolates. The category "RIII" with sextuple mutation in two locus combination suggested for level of resistance that can lead to total failure against SP treatment was observed as genotype GEN12-13 in 3.22 % (n = 7) isolates.

Discussion
Currently, combination of fast and long acting antimalarial drugs is recommended as an ideal approach over the use of single antimalarial drug [3]. Optimizing the choice of long acting partner antimalarial drug in ACT is important challenge to be addressed in successful malaria treatment programme [33]. Presence of resistant parasites against the long acting antimalarial used in ACT can hamper the treatment efficacy and can also lead to emergence of artemisinin resistant parasite [7]. All malaria endemic parts of India experienced mutant parasites conferring resistance to all conventional antimalarial drugs like CQ, SP and thus there was country wide adoption of AS + SP as ACT in year 2010 [5]. However, resistance to SP had been well documented from northeastern part of India, which led to use of AS + lumefantrine as first-line malaria treatment in these parts of country since year 2013 [7]. Northeast region of India has already been documented as important route for invasion of parasite bearing resistant genotypes against many antimalarials and proved its potential to be a focus for spread of resistant parasite to other parts of country [20]. Thus, monitoring of mutation status of partner SP is important for better management of antimalarial policy. Here, mutation status of pfdhfr and pfdhps gene responsible for resistance against SP was evaluated for isolates from four different geographic areas.
The study showed 17.51 wild-type pfdhfr gene and 79.26 % wild-type pfdhps gene. Higher number of mutant pfdhfr gene was observed in comparison to pfdhps gene at all the study sites infer development of mutations occurred first in pfdhfr gene and then in pfdhps gene under selective drug (SP) pressure. The prevalence of double mutant (ANRNI) in pfdhfr gene and wild-type pfdhps gene at all the study sites corroborated earlier reports of predominant presence for the same [23]. However, single mutant ANCNI, triple mutants (ANRNL or AIRNI) in pfdhfr gene and single mutants (SGKAA, AAKAA and SAEAA), double mutant SGKGA, triple mutant AGEAA, quadruple mutant AGEGA were also observed. Single or double pfdhfr mutations alone cannot cause SP treatment failure but the double pfdhfr mutations with a single or more mutation in pfdhps gene can cause various level of SP resistance [15]. In addition, triple mutant pfdhfr alone can cause various level of SP resistance. The DHFR-DHPS two locus mutations have importance to monitor as it can infer the clinical susceptibility of SP [15,18]. This study observed 13 such two locus genotypes (GEN 1-13) within 217 isolates ( Table 4). Out of all isolates, only 17.51 % were wild-type (GEN1). In total, double mutant genotype (GEN3) was observed in 52.99 % isolates and its predominance indicates continuous emergence of SP resistance in all study sites. The study sites include both high malaria transmission area (Odisha and Jharkhand) and low malaria transmission areas (Andhra Pradesh and Uttar Pradesh). Triple mutant genotype (GEN 4-8) that can confer high SP resistance were observed in Odisha and Jharkhand with prevalence of 12.85 and 5.95 % respectively, while quadruple mutant (GEN9) was found only in Jharkhand with 21.42 % prevalence. Isolates from high transmission areas also showed quintuple (Jharkhand = 5.95 %, Odisha = 11.43 %) and sextuple (Jharkhand = 4.76 %, Odisha = 4.29 %) mutant genotype. Quintuple and sextuple mutant genotypes associated with higher level of resistance to SP suggests selective drug pressure due to its use since a long period. Here, high transmission areas showed higher number of mixed mutations (both wild and mutant alleles) in both pfdhfr and pfdhps genes, which were possibly due to multi-clonal infection, as high recombination event is expected here, which in turn adding allelic variation. High genetic diversity at these high transmission areas in both gene under selection and neutral microsatellite markers were reported when compared to low transmission region of India [34,35]. Higher genetic diversity and more fixation probability in genes responsible for various antimalarial resistance was observed earlier and suggested the role of malaria transmission intensity and drug exposure [36,37]. The mutations 164L in pfdhfr and 437G and 540E mutations in pfdhps gene were reported to be responsible for therapeutic failure of SP [20,22], and the same were observed in 9.52 and 34.52 % respectively in Jharkhand. In Odisha 4.28 % isolates showed another mutation 51I in pfdhfr gene which was responsible to accelerate the SP-resistance [15,16]. In addition, these mutations are also part of two-locus genotypes (GEN 7-13) would be involved in clinical resistance against SP. In pfdhps gene, triple mutant AGEAA was found in 15.71 % of Odisha isolates, while double mutant SGKGA is found only in 23.80 % of Jharkhand isolates. The prevalence of mutants found here in high transmission areas are similar to reported earlier from northeastern region [20,22], however the prevalence of mutant two locus genotypes were not similar. Mutations like S436F, A613T/S was not observed in this study.
In contrast, the low transmission areas (Uttar Pradesh and Andhra Pradesh) showed single mutation (11.11 %) at codon positions 108, double mutations (66.67 %) at codon positions 59 and 108 in pfdhfr gene, while no isolate showed the N51I and I164L mutations associated with SP treatment failure. Thus the triple and quadruple mutations were not observed in pfdhfr gene. In case of pfdhps all isolates were wild-types, which infers P. falciparum population in these regions were susceptible to SP treatment and resistance could be in developing state. The low transmission areas showed mutations similar to earlier reports of similar single and double mutations from Uttar Pradesh and Delhi in year 2001, which suggested higher susceptibility for SP was maintained due to higher clonal populations in these regions [18]. In addition, P. vivax is prevalent in Uttar Pradesh and chloroquine is still effectively used as antimalarial treatment against P. vivax in India, which could provide selection pressure on gene responsible for chloroquine resistance in P. falciparum [34]. Thus no or low selection pressure of antifolate drugs in P. falciparum was predicted in these P. vivax prevalent areas as misdiagnosed of mixed infection cases was more exposed to chloroquine in comparison to antifolate drug .

Conclusion
In conclusion, the present findings suggest that SP can be effective for the treatment of uncomplicated falciparum malaria as a partner drug of ACT in Andhra Pradesh and Uttar Pradesh (low transmission areas). In Jharkhand and Odisha (high transmission area), results suggest that mutation rate will increase continuously due to continued drug pressure and malaria transmission, which in turn will lead to SP treatment failure in near future, as reported in northeastern parts of India. Continuous molecular surveillance of partner drug (SP) in these high transmission areas is required to maintain an effective drug policy.