Genetic profiling of the Plasmodium falciparum parasite population in uncomplicated malaria from India

Background The genetic complexity and the existence of several polymorphisms in parasites are the major hindrances for the malaria control programmes of the country. The genetic profiling in the parasite populations in India will provide useful baseline data for future studies elucidating the parasite structure and distribution of drug resistance genotypes in different regions. Methods The blood samples of symptomatic patients were collected and analysed for drug resistance genes (Pfcrt, Pfmdr-1, dhfr, dhps and k13) and gametocyte genes (Pfs25, Pfg377); in vitro drug sensitivity assay by schizont maturation inhibition (SMI) was also performed in adapted field isolates. Results Of the 122 field isolates analysed; 65.5% showed Pfcrt K76T mutant alleles, 61.4% Pfmdr-1 N86Y mutants, 59.5% dhfr mutants, 59.8% dhps mutants was observed, but no polymorphism was seen for k13. The sequence analysis of Pfg377 gene revealed five types of populations in the field isolates. The inhibitory concentrations (IC50) for anti-malarial drugs viz chloroquine (CQ), artesunate (AS), were in the range of 10.11–113.2 nM and 2.26–4.08 nM, respectively, in the field isolates evaluate by in vitro assay. The IC50 values for CQ have shown a remarkable reduction on comparison with the previous available data, whereas a slight increase in the IC50 values for AS was observed in the study. Conclusions The increase in mutation rate in drug resistance allelic loci with inhibitory concentration of CQ and AS drugs was observed in the field isolates and high diversity in Pfg377 gametocyte gene indicate towards parasite multifactorial behaviour. The knowledge of the prevalent drug resistance genes is important for intervention measures to be successful and efforts should also be made to prevent transmission of P. falciparum.


Background
Morbidity and mortality due to Plasmodium falciparum malaria have increased across Indian sub-geographical regions, largely because of the widespread resistance to chloroquine (CQ) and sulfadoxine-pyrimethamine (SP) leading to generation of regional heterogeneity in the target genes [1,2]. The rise in the incidences of malaria is largely attributed to the increase in the drug resistance parasites [3]. The emergence of chloroquine resistance in India led to the change in the drug policy in the year 2005 to artemisinin-based combination therapy (ACT) as first-line of treatment [4]. The Pfcrt and Pfmdr-1 genes are associated with CQ resistance, whereas the dhfr and dhps genes are related with SP resistance in the malaria parasite caused by the mutations present in these drug resistance genes [5]. The Kelch propeller domain (k13-propeller) in P. falciparum have been found to be associated with delayed parasite clearance after ACT therapy [6].
Plasmodium falciparum is transmitted from the human host to the mosquito vector during the sexual stage via gametocyte. But the transmissibility of P. falciparum is due to the relationship between the prevalence, duration and density of gametocyte carriage with variable and modulated by the immune response in the human host [7]. Gametocytes from multi-clone P. falciparum infections can persist three times longer than singleclone infections indicating that multiplicity of infection (MOI) promotes either longer persistence or continuous production of gametocytes [8]. The occurrence of sub-microscopic gametocytes in the parasite population play a major role in malaria transmission. It has also been shown that the microscopic diagnosis is more likely to underestimate the prevalence of gametocytes and more so in the sub-patent population of gametocytes [9]. For the successful elimination of P. falciparum it is crucial to limit the transmission of malaria parasites which can be achieved by reducing the carriage of viable gametocytes, hence it is important to investigate the drug resistance genes with gametocyte genes [9].
The aims of the present study were to determine the prevalence of Pfcrt, Pfmdr-1, dhfr and dhps mutations in drug resistance genes with k13 (particularly ACT with SP as partner drug and CQ) and Pfg377 genes with in vitro susceptibility assays in order to understand the genetic complexity of falciparum malaria in natural infections. The Pfcrt, Pfmdr-1, dhfr and dhps are directly implicated in parasite population turning resistant to anti-malarials and Pfg377 gene is considered under neutral/balancing selection. Coupling the analysis of drug resistance genes along with k13 with in vitro studies will give an insight about the circulating parasite population in the region and profiling the chloroquine markers will provide data to know if there is any change at the genetic level for these markers after a change in the treatment policy.

Study population and ethics
In the present study, 143 malaria blood samples were collected from the primary health centres (PHCs) in Mewat, Raipur and Rourkela during the transmission season in the years 2015-2016 (Fig. 1). The study group comprised of patients of either sex, 5-60 years of age, with symptoms of malaria, but no history of prior medication until previous 30 days. Two microliter of venous blood from patients found to be P. falciparum positive by microscopy/RDT were collected in EDTA vacutainer tubes. 100 µl of the collected sample was used for filter paper spots, labelled and transported in the laboratory in sealed plastic envelopes and remaining samples were cryopreserved.
The institutional ethics committee (NIMR-IEC) approved the study (reference no. ECR/NIMR/EC/2) and the ethical guidelines of the Institute were followed for sample collection. Verbal and written informed consent was obtained from all patients/guardians in case of minors.

Nucleic acid extraction
The parasite genomic DNA from filter paper was extracted using QIAamp TM DNA mini kit (QIAGEN), according to the manufacturer's instructions (dried blood spots protocol) with slight modifications. DNA samples were kept at − 20 °C until further processing.

PCR analysis
PCR assay for Plasmodium species: After the DNA extraction nested PCR assay with 18S rRNA primers was performed to differentiate between the mixed infections of Plasmodium vivax and P. falciparum, as described previously [10].

PCR-RFLP assay
After successful PCR amplification the polymorphisms in Pfcrt and Pfmdr-1 genes were determined by RFLP assay by digestion with ApoI and AflIII enzymes. The enzyme ApoI cleaves the Pfcrt-76K (the CQ sensitive isolates) but not Pfcrt-76T (as seen in CQ resistant isolates) and allele of Pfmdr-1-86N (the CQ sensitive isolates) but not Pfmdr-1-86Y (the CQ resistant isolates). AflIII was unable to digest the amplicon of the sensitive isolates.

Sequence analysis
The k13 amplification reactions were carried out as described earlier and mutations were assessed by comparing each sequence with the reference 3D7 K13-propeller (PF3D7_1343700) [6]. The positive PCR products of dhfr and dhps genes were purified and sequenced on an ABI 3730 DNA analyzer. The 3D7 strain (GenBank ID: for dhfr 9221804 and dhps 2655294) was used as reference for comparing the gene sequences. The sequence comparisons was done with the Plasmo database of these two genes with basic local alignment search too N (BALST N) and MEGA software [13].

Parasite culture
In the laboratory, the cryopreserved samples were revived in the RPMI 1640 culture medium supplemented with gentamicin (0.01 mg/ml), 25 mM Hepes buffer, 25 mM NaHCO 3 and 5% human serum maintained in 5% CO 2 with incubation at 37 °C. The growth of the parasite in vitro was visualized daily by microscopic examination of the Giemsa-stained blood films [14].

In vitro analysis
The drug plates were prepared in pmol/l as per the WHO in vitro microtest procedure by SMI assay [15]. CQ, AS (Sigma-Aldrich) stock solutions of 10 mM concentrations were diluted two folds to coat the drug plates [16]. A 50 μl of the blood media mixture (BMM) was added in the pre-dosed drug plate in triplicate with serial drug dilutions and then incubated in CO 2 gas mixture at 37 °C for 24-30 h for schizont maturation. After 24 h incubation, a thin smear was prepared from the control well to observe mature schizonts and if more than 10% schizonts (with more than 3 nuclei) were seen in the control wells, it was considered to be valid. The thick smears were Giemsa-stained and the schizonts were counted out of a total of 200 asexual parasites for calculating IC 50 values. For the test, the 3D7 was used as the reference strain with a mean 19.1 nM and 2.4 nM IC 50 value for CQ and AS, respectively.

Detection of gametocyte genes
Pfs 25 and Pfg377gene: The PCR analysis for Pfs25 and Pfg377 genes in the patient samples were carried out using the PCR conditions with high fidelity Taq DNA polymerase as described previously (ThermoFisher Taq) [16,17].

Sequence analysis
The nested PCR products for region 3 of Pfg377 were purified and sequenced using the cycle sequencing reaction with ABI Big Dye terminator version 3.1. The Pfg377 gene of 3D7 strain (genID: Pf 3D7_1250100) was used as reference for sequence comparisons of the field isolates. The sequences were then assembled, aligned and edited using MEGA 6 and Finch TV program [13].

Statistical analyses
The aligned dhfr, dhps and Pfg377 gene sequences were used to determine the number of haplotypes. Also ɵ and π, the two measures of nucleotide diversities were estimated for each parasite population in each region [18]. The π nucleotide diversity parameter is based on the average number of pairwise nucleotide differences per site and ɵ estimates is dependent on the number of segregating sites. All these were estimated using DnaSP version 5.10.01 [18].

Samples
Out of the total 143 samples analysed for mixed infections of Plasmodium spp. six samples were found to be of mixed infections with P. falciparum and P. vivax. Only P. falciparum mono-infections 122 sample (single parasite infections) were analysed in this study and P. vivax and co-infections were not further proceeded for any other examination. No follow up study was done to evaluate the parasite clearance in the patients hence the clinical data on delayed asexual stage parasite clearance was not available. Table 1 gives the distribution of these genes in the three studied population.

Pfcrt, Pfmdr-1 gene alleles and k13
Pfcrt gene was examined for K76T allele and was successfully amplified in 122 of field isolates. After digestion with ApoI enzyme the mutant allele 76T was seen in 80 (65%) in samples and 42 (37.4%) were wild type isolates indicating their sensitivity to CQ ( Table 1).
The Pfmdr-1 gene amplicons were digested by restriction enzyme AflIII. Upon digestion, different patterns of N86Y were observed where wild N-codon was in 47 isolates (38.5%) wild and mutant Y-codon in 75 isolates (61.4%) and mixed N/Y in two isolates (2.46%). We did not observe any significant correlation between the Pfcrt K76T and Pfmdr-1 N86Y SNPs in the study ( Table 1). The k13 amplified did not reveal any synonymous or non-synonymous prevalent mutations in the field isolates.

dhfr and dhps genes
The dhfr and dhps genes for SP were sequence analysed for the presence of mutations in 122 and 121 isolates respectively. Different isolates showed presence of different SNPs and a rise in double and triple mutations was observed in both the genes ( Table 2). In dhfr gene the mutant 59R associated with high-level SP resistance was found to be 42.6% and for codons 51I, 108N and 164L mutants were observed at 31.9%, 17.2% and 7.3% in the population respectively. The wild genotype for dhfr was seen at higher frequency in Raipur with 37.5% followed by 27.2% in Ranchi and 22.9% in Mewat regions.
The dhps gene mutations at codon 436F was 47.4% and 540E was 42.6% respectively whereas 437G, 581G and 613T occured at 7.3%, 5.7% and 1.6%, respectively. In dhps gene the wild type field isolates were highest in Mewat region with 23.7% followed by 13.1% in Raipur and 3.2% least in Ranchi ( Table 2). The triple mutation associated with high level SP resistance 51I R59 N108 was highest in Mewat followed by Raipur and Ranchi with 36%, 35.4% and 20.4%, respectively.

Genetic diversity analysis
The sequence analysis of dhfr and dhps genes revealed highest nucleotide diversity in Ranchi for both the genes where the ɵ and π estimates of P. falciparum isolates showed an increase in the nucleotide diversity on comparison with Raipur and Mewat ( Table 2). The π values suggest slight increase in intermediate frequency mutations in the population. The neutrality tests show no statistically significant results for the tests carried out for any divergence in both genes in the study population (Table 2).

Sequence analysis
The region 3 of Pfg377 gene was included in the study as it is the most polymorphic region and hence used in allele typing. Sequencing the 3 regions of the Pfg377 gene was carried out for 122 isolates revealed the deletion of 21 bp and multiples of 21 bp. On comparison with NF54 reference sequence, the isolates showed seven different types of deletions viz NHHIDHQ, HNHHIDH, HNHHIDHHNHHIDH, DHHIDHH, NHHIDHH, NHHIDHHNHHIDHQ and NHHIDHHNHHIDHH-DHHIDHHDHHIDHH amino acid fragments. The most commonly occurring deletion was HNHHIDH, found in 42 isolates. The isolates with no deletions exhibited DHHI, DHH and NHHI sequence repeats 4, 9 and 7 times, respectively. These repeats occurred at different frequencies in the other forty isolates repeating these sequences 4, 8 and 6 times, respectively. On the basis of these deletions five types of genotype was present in the field isolates and He for Pfg377 gene was found to be 0.75. Two new mutations-D924N and H888Q mutation was found in four and nine isolates respectively on sequence analysis (Fig. 2).
Combining the mutational analysis of the dhfr, dhps and Pfg377 genes 18 haplotypes were observed in the present study in all the populations (Table 3). Among these haplotypes most prevalent was A10 commonly seen in Raipur and haplotype A18 depicted the wild population (n = 44) seen distributed randomly in Mewat (45%), Raipur (37.5%) and Ranchi (9%). The combined results of dhfr and dhps mutational analysis reveals highest polymorphisms in parasite populations from 77.7 and 81.8% respectively in Ranchi.

Phylogenetic analysis of dhfr, dhps and Pfg377 genes
To identify the existing genetic diversity in the parasite population the radial cladogram network was constructed between the drug resistance loci dhfr, dhps and gametocyte gene Pfg377 by using Dendroscope version 3.5.9 software [19]. The genetic relatedness is determined by clade in each phylogeny tree which signify the similarity in the various isolates. In dhfr phylogeny tree we find that the isolates were divided in two clades: the first clade was formed by the wild type isolates along with the reference strain and the second clade comprises of isolates having mutant alleles. The mutant clade consists of different sub-group clades within isolates having single, double or triple mutations (Fig. 3a). The sub-group clade subdivisions in different colours depicts the different mutations in the populations. The most predominant mutation clade in the cladogram consists of 59 and 108 codons while the least prevalent 108 mutations made

Table 2 Nucleotide diversity in dhfr and dhps allelic loci in sub-population areas with their different test parameters
Tajima's D, Fu, Li's D* and Li F* test used for standardized measure of average number of mutations between pairs in the sample. Nucleotide diversity (Ɵ) and pairwise difference (π) indicates the distribution of mutation rate which varies among the regions. # Total number of field isolates in the parenthesis for that particular region. P value for all the three sites for the two genes was found to be non-significant (< 0.05) the other clade. The same sequence relatedness was also found in dhps allelic profiling (Fig. 3b). The most predominant in this allelic profile was 436 followed by 540 codons and the least one was codon 613 position. In Pfg377 the clades are similar with variations due to short sequence deletions and repeats among the population (Fig. 3c). The gene distribution shows no particular pattern for any mutant alleles of dhps and dhfr distributed across the country and to any particular region (Fig. 3a,   b). Similarly, for Pfg377 gene too no distinct pattern for deletions or repeats were seen, with all randomly distributed in the population (Fig. 3c).

Culture studies
The cryopreserved samples were subjected to in vitro culture for revival and 44 isolates adapted in culture. After successful establishment of culture adapted  samples, the isolates were put for gametocyte production in vitro. After 15 days of culturing in laboratory, 10 field isolates produced gametocytes and various stages I-V were identified by microscopy. From 7th day onwards post sub-culture, different stages of gametocytes were seen in smears. In two isolates no gametocyte production was seen though the asexual stages were identified in smears during culture adaptation in laboratory.

In vitro drug sensitivity assay
Due to limited volume of cryopreserved samples it was technically difficult to analyse with all anti-malarial drugs hence only primary drugs-CQ and AS in vitro sensitivity assay was performed using the SMI assay. The CQ resistant isolates showed schizont maturation at 8 pmol or more and for AS no resistance was seen in isolates.
The inhibitory concentrations (IC 50 ) for the CQ and AS, drugs were found in the range of 10.11-113.2 nM, 2.26-4.08 nM, respectively in the evaluated field isolates (Fig. 4a). The IC 50 values for CQ have shown a remarkable reduction on comparison with the previous available

Discussion
The malaria situation continues to prevail even after intensive control efforts and remains persistent in India; as India is a major hot-spot for malaria transmission due to its unique favourable topography with respect to malaria survival [20]. Drug resistance is a major hurdle in malaria control not only in the country but worldwide too [21]. Continuous monitoring of the drug resistance provides warning signals to drug policy makers. The emergence of drug-resistant parasites in India is still on the rise even after the introduction of ACT in the country [4]. Genotyping of Pfcrt, Pfmdr-1, dhfr, dhps genes and k13 was carried out in field isolates of P. falciparum collected from Ranchi, Raipur and Mewat to determine the prevalence of polymorphisms in these genes along with the neutral gametocyte gene viz Pfg377. Not many such studies have been reported from India. In the study, non-drug resistant Pfg377gene showed lot of genetic diversity in the field isolates which agrees with the observed random association between the loci and absence of regional differentiation among the P. falciparum isolates is in accordance with other reports [22,23]. CQ resistance (CQR) was monitored on the point mutation detected in the Pfcrt and Pfmdr-1 genes by PCR-RFLP method where a high prevalence of mutant K76T Pfcrt and N86Y Pfmdr-1 alleles was observed. This holds true with the previous studies stating the role of Pfcrt gene crucial in detection of CQR in P. falciparum field isolates [24][25][26]. No linkage disequilibrium (LD) between the N86Y and K76T mutations in Pfmdr-1 and Pfcrt genes was observed and more number of samples are to be studied to compare the LD association in P. falciparum isolates. It is extremely important to routinely detect/analyse the pattern of drug resistance in malarial parasites which play not only an important role in the epidemiological surveys but also help in improvising/ updating the anti-malarial drug policies of the country. The CQ resistance if P. falciparum is mainly linked to the mutation in K76T codon of Pfcrt gene followed by Pfmdr-1 gene N86Y [24,25].
SP are the partner drugs in the current ACT therapy of the country and we observed a high prevalence of more than one mutations in the dhfr, dhps genes of P. falciparum [27]. The double mutants 51I 108N in the dhfr gene were present at a higher frequency in the population as compared to the triple mutant 51I 59R 108N associated with high resistance for SP was only seen in the field isolates from Mewat [28]. It has been reported from Africa and South East Asia earlier that I51 R59 N108 are highly SP resistant haplotype and the presence of this genotype in the studied population reveals that the parasite population is possibly moving towards SP resistance where evidence of selection is seen earlier in dhfr gene than dhps gene [29,30]. This indicates the possibility of high resistant genotypes evolving in the parasite population due to constant drug pressure experienced by the parasite [31]. The π values of nucleotide diversity were higher for dhfr and dhps genes in Mewat and similar in Ranchi and Raipur the endemic regions for malaria similarly documented from other places [32]. Analysing the SNPs in these drug resistance genes to find new mutations will help us to understand genetic changes leading to altered phenotypic traits better and also give an insight into the evolving genetic recombinations with respect to the current drug therapy [33]. The usage of ACT in the national drug policy since a decade is expected to decrease the CQ resistance and restore the CQ sensitive parasite population as such a change has been reported from several countries and 31.8% wild type of isolates in the parasite population was also seen in the present study [34].
Similarly, though the majority of isolates were found to be resistant to CQ, the IC 50 observed value for CQ was comparatively low as seen in other reports [35]. This low value is indicative of the fact that the absence of CQ in malaria treatment has released the drug pressure with a possibility of moving towards being CQ sensitive in future as reported from other parts of the world [36]. The field isolates were found to be AS sensitive through with high IC 50 values. Due to constant exposure to AS the increase in the drug pressure is likely to be responsible for increase in the IC 50 values and another plausible explanation also could be the indiscriminate use of the drug resulting in slow rise of low level resistance in the parasite population [37]. In vitro drug sensitivity remains an important tool to assess the efficacy of the antimalarial drugs. The change of anti-malarial treatment in the country from CQ to ACT is expected to decrease the CQR population and might be leading to restore the sensitivity to CQ as reported from other places [38]. The levels of parasitaemia varied in the infections and only 44 samples could be culture adapted. The difference in the growth pattern in different field isolates may be due to their difference in commitment to differentiate into asexual or sexual stages [39]. The expression profile of Pfg377 gene was correlated with the efficiency of gametocyte production in the above isolates. It was observed that isolates which produce mature gametocytes in vitro also showed an increase in thePfg377 gene expression from 1.00 to 4.56-fold when compared to NF54 reference strain. The results suggested that a correlation between Pfg377 gene expression and the ability to produce gametocyte (mature) by the isolate exists as also reported by others [40].
The allele typing of Pfg377-the sexual stage specific gene in the field isolates revealed several allele types present in them. By conventional PCR only the predominant genotype was detected in the PCR results and presence of more than one clone in a single infection were observed on sequence analysis. This shows that this gene can be effectively used for detecting the presence of gametocyte producing multiclones in the field isolates. The presence of different genotypes/clones in the same infection is an important source of transmission [8]. The large number of repeats found in this gene is responsible for the high variability seen in parasites thus allowing several parasite lines to merge and survive simultaneously leading to transmission dynamics [41]. It is understood that presence of mixed gametocyte genotypes in the same sample is responsible for harbouring gametocytes for a significantly longer period of time on comparison with a single genotype where it was cleared three times faster than multiple clones [42]. Natural infections of P. falciparum seem to harbour multiple genotypes in the same host which could be due to the transmission burden in endemic regions but still, this knowledge does not improve the understanding of existing multiplicity in parasite biology [43].
By substantiating the results obtained from the genome sequence of P. falciparum and phylogenetic analysis, a basic idea about the population genetic structure was inferred. The phylogenetic trees of all three genes divided the isolates into two clear clades that can potentially be divided into further subgroups. However, no obvious cluster was observed and the distribution of isolates within the subgroups was random. There was no apparent trend for sequences of the same geographic region to be more closely related. The non-availability of wild-type of isolates limits our understanding of wild type parasites vs drug-resistant parasites as most of the infections had at least single/double SNPs in drug resistance genes with extensive diversity present in the Pfg377 gene. More comprehensive studies like this are needed to understand the emerging complexity in the parasite population. The gametocyte data in relation to diversity level is too limited in Indian context. Our main focus to include this data was to generate data for any type of correlation between the mutation in drug resistance genes and gametocyte gene as these gametocyte genes play a crucial role in transmission as it gives baseline data for further work in this area.

Conclusion
This study provides information on drug resistance phenotypes, which have a critical role to play in malaria transmission. The increasing mutation rates in dhfr, dhps genes, existing diversity in the Pfcrt, Pfmdr-1 and Pfg377 genes, increase in the IC 50 values for AS, in the parasite population indicates that as the parasite genetic diversity is multifactorial; the emergence of drug resistant genotypes need to be tackled by multipronged approach. The knowledge of the host dynamics with its epidemiological and evolutionary consequences is important for intervention measures to be successful and efforts should also be made to prevent transmission of P. falciparum. To assess the resistance dynamics and the surveillance of molecular markers of drug resistance is essential as it helps in selecting the most effective treatment and also should help in malaria intervention programmes.