Extensive Genetic Diversity in the Plasmodium falciparum Na⁺/H⁺ Exchanger 1 Transporter Protein Implicated in Quinine Resistance

Abstract

The Plasmodium falciparum Na⁺/H⁺ exchanger (Pfnhe-1) locus at chromosome 13 and another locus at chromosome 9 have recently been proposed to influence quinine resistance. Here, we sequenced the ms4760 locus of the Pfnhe-1 gene from 244 P. falciparum isolates collected from five different regions of India. A total of 16 different ms4760 alleles (with one to five DNNND repeats) were observed among these isolates. Interestingly, areas with a high prevalence of chloroquine and sulfadoxine-pyrimethamine resistance showed more Pfnhe-1 DNNND repeats compared to low drug resistance areas. The extent of genetic diversity at the ms4760 locus also varied from one region to another, with expected heterozygosity values ranging from 0.47 to 0.88.

The emergence and spread of resistance in Plasmodium falciparum against the two most commonly used antimalarial drugs, chloroquine (CQ) and sulfadoxine-pyrimethamine (SP), has resulted in an increased use of quinine (QN) (6, 11). This is because QN is not only used to treat severe and complicated malaria cases but is also used to treat patients who do not respond to CQ and SP treatment. While the genetic basis of CQ and SP resistance in the parasite is known in great detail (17), the same is still evolving for QN resistance (7, 8). The possible involvement of the P. falciparum Na⁺/H⁺ exchanger (Pfnhe-1) membrane transporter at chromosome 13, along with another locus at chromosome 9, in QN resistance has recently been proposed (8, 9). We describe here the genetic variation in the cytoplasmic domain of the Pfnhe-1 protein containing DNNND repeats (ms4760 locus) among Indian isolates.

The P. falciparum isolates used in this study were collected during 2004 and 2005 from Panna (Madhya Pradesh [MP]; n = 45), Ghaziabad and Aligarh (Uttar Pradesh [UP]; n = 50), Car Nicobar (Andaman and Nicobar [A & N] Islands; n = 55), Kamrup (Assam; n = 74), and Cuttack (Orissa; n = 20) (Fig. 1). Patients attending malaria clinics were screened for the presence of malaria parasites by light microscopy and treated with antimalarial drugs as described earlier (4), in accordance with the national drug policy of India prevalent in the region (http://www.nvbdcp.gov.in). About 200 μl of heparinized blood was collected from P. falciparum-positive individuals. Blood collections were done in accordance with institutional ethical guidelines. Parasite DNA was extracted in accordance with the protocols described earlier (18). An aliquot was used to amplify the ms4760 locus of the Pfnhe-1 gene with primers NHE-A (5′-AGTCGAAGGCGAATCAGATG-3′) and NHE-B (5′-GATACTTACGAACATGTTCATG-3′) and standard PCR cycling parameters (18). Primers NHE-C (5′-ATCCCTGTTGATATATCGAATG-3′) and NHE-D (5′-TTGTCATTAGTACCCTTAGTTG-3′) were used for nested PCR. Both strands of the PCR product were sequenced with primers NHE-C and NHE-D by the protocols described earlier (18).

FIG. 1.

Map of India showing the sample collection sites.

Sequencing of the ms4760 locus of the nhe-1 gene from 244 P. falciparum isolates revealed extensive polymorphisms at this locus (Fig. 2) with an overall calculated heterozygosity (He) value of 0.79. The level of genetic variation among the isolates from UP and MP was relatively lower than that of the isolates from Assam, Orissa, and the A & N Islands (Table 1). A total of 16 different ms4760 genotypes of Pfnhe-1 were observed among these isolates. This includes 5 of 8 already reported Pfnhe-1 genotypes besides 11 newer genotypes (8). The frequency of occurrence of these genotypes varied among the regions (Table 1). Genotypes ms4760-2, ms4760-4, and ms4760-8, reported by Ferdig et al. (8), were not detected among our Indian isolates. Genotypes ms4760-3 and ms4760-6 were present in all five regions (Table 1). A maximum of 12 genotypes were found in Assam, while UP had only 3 genotypes.

FIG. 2.

Multiple amino acid sequence alignments of 19 ms4760 genotypes. The genotypes ms4760-2, ms4760-4, and ms4760-8 are from Ferdig et al. (8), while the rest of them are from this study. Block I represents a DNN repeat, block II represents DNNND repeats, block III is characterized by insertion and deletion of NHND residues, block IV shows insertion and deletion of the amino acids DKNNKND, block V represents DDNHNDNHNND repeats, and block VI represents DDNNNDNHNDD repeats. All other variations are highlighted. Gaps were created in order to maximize the alignment. The designations on left are microsatellite genotypes, and the numbers on right are positions of amino acids. The electropherograms were analyzed with the BioEdit Sequence Alignment Editor (10), and sequences were aligned with the GeneDoc Multiple Sequence Alignment Editor and Shading Utility (version 2.6.002).

TABLE 1.

Regional distribution of Pfnhe-1 ms4760 genotypes among Indian isolates

Genotypea or parameter	No. of DNNND repeats (His/Asp ratio)b	No. of isolates (allelic frequency)					Total (n = 244)
		MP (n = 45)c	UP (n = 50)	A & N (n = 55)	Assam (n = 74)	Orissa (n = 20)
ms4760-1	2 (7:4)	3 (0.066)		4 (0.072)	6 (0.081)	3 (0.15)	16 (6.55)
ms4760-3	1 (6:2)	7 (0.155)	17 (0.340)	12 (0.218)	8 (0.108)	1 (0.05)	45 (18.44)
ms4760-5	4 (5:8)	1 (0.022)		2 (0.036)		1 (0.05)	4 (1.630)
ms4760-6	2 (5:4)	32 (0.711)	28 (0.560)	6 (0.109)	16 (0.216)	6 (0.30)	88 (36.06)
ms4760-7	3 (5:6)			27 (0.490)	17 (0.229)	3 (0.15)	47 (19.62)
ms4760-9	3 (7:6)	2 (0.044)		4 (0.072)	8 (0.108)	2 (0.10)	16 (6.550)
ms4760-10	1 (7:2)				2 (0.027)		2 (0.810)
ms4760-11	3 (5:6)				5 (0.067)	1 (0.05)	6 (2.450)
ms4760-12	1 (8:2)				7 (0.094)		7 (2.860)
ms4760-13	2 (5:4)				1 (0.013)		1 (0.409)
ms4760-14	3 (6:6)				1 (0.013)		1 (0.409)
ms4760-15	3 (5:6)				2 (0.027)		2 (0.810)
ms4760-16	2 (8:4)		5 (0.100)				5 (2.040)
ms4760-17	5 (5:10)					1 (0.05)	1 (0.409)
ms4760-18	2 (7:4)				1 (0.013)	1 (0.05)	2 (0.810)
ms4760-19	1 (6:2)					1 (0.05)	2 (0.409)
Hed		0.47	0.57	0.70	0.86	0.88	0.79

^aThe ms4760-2 (one DNNND sequence), ms4760-4 (three DNNND repeats), and ms4760-8 (three DNNND repeats) genotypes reported elsewhere (8) were not found in this study.

^bHis/Asp ratios of ≥1 are in boldface type.

^cn, number of isolates tested.

^dThe expected He was calculated by using the formula He = [n/(n − 1)](1 − Σp_i²), where n is the number of samples genotyped for that locus and p_i is the frequency of the ith allele. Statistical analysis was done with the chi square test, and a P value of <0.05 was considered significant. DNA sequence polymorphisms (nucleotide diversity per site; haplotype [gene] diversity and minimum number of recombination events between sites) were estimated with DnaSP ver. 4. 10. 9 (15).

The number of DNNND repeats varied from one to five, and two repeats were more common (45.9%, n = 244) among the isolates (Fig. 3A). The distribution of isolates with different DNNND repeats exhibited a regional bias (Fig. 3B). Significantly larger numbers of isolates from the A & N Islands, Assam, and Orissa (areas of high CQ and SP resistance) were found to contain a higher number of DNNND repeats compared to those from UP and MP (areas with relatively low CQ and SP resistance; 1-3, 13, 14). However, we have observed that isolates with increased numbers of DNNND repeats also contained a higher number of mutations in the P. falciparum CQ resistance transporter (P < 0.05) (data not shown). This indicates that there could be an association or relationship between the mutations of these two genes. The exact reason for this geographic variation in Pfnhe-1 is not known and requires further investigation. Earlier, Ferdig et al. (8) found a statistically significant association between the number of DNNND repeats and the level of QN resistance. Although it may not directly correlate with the clinical outcome, the number of DNNND repeats may indirectly influence QN responsiveness, as it can influence the His/Asp ratio of the C-terminal domain of Pfnhe-1 (7). The His/Asp ratio can affect the cytosolic pH of the parasite through interactions with other loci and thus the exchange of Na⁺ and H⁺ and probably the QN response (7). The example shown in the literature is the ms4760-1 genotype-containing parasite showing a higher level of QN resistance, as this genotype contains a higher His/Asp ratio (7:4) compared to the QN-sensitive parasite bearing the genotype ms4760-5 with a His/Asp ratio of 5:8 (7, 8). In this context, only 6.5% (n = 244) of the Indian isolates contained the ms4760-1 genotype (Table 1). Indeed, another genotype, ms4760-18, was also found to contain the same His/Asp ratio, which was present in two isolates only. Thus, there were a total of 18 (7.38%, total n = 244) isolates with a 7:4 His/Asp ratio, which could follow the pattern of QN responsiveness described by Bennett et al. (7).

FIG. 3.

Distribution of DNNND repeats among all of the Indian P. falciparum isolates used in this study (A) and regionwise (B). n, number of isolates.

Our findings of more allelic diversity among the isolates from Assam (12 genotypes, He = 0.86), Orissa (10 genotypes, He = 0.88), and the A & N Islands (6 genotypes, He = 0.70) compared to the isolates from MP (5 genotypes, He = 0.47) and UP (3 genotypes, He = 0.57) are in agreement with reports on the mutations in the markers associated with other drug resistance in these areas (2, 3, 14). Besides, the rates of malaria transmission are also higher in Assam, Orissa, and the A & N Islands compared to those in MP and UP (16). Furthermore, the degree of He (∼0.79) at the ms4760 locus of the Pfnhe-1 gene is also comparable to the He observed at other microsatellite loci in P. falciparum (5, 12, 18). From a total of 378 nucleotide positions considered in the alignment, 42 were polymorphic (18 singleton variable and 24 parsimony informative) and 178 were monomorphic sites. The haplotype (gene) diversity was 0.90 ± 0.062 (standard deviation). The nucleotide diversity per site was found to be 0.0515 ± 0.0081, and the nucleotide diversity with Jukes-and-Cantor correction was 0.054. The minimum number of recombination events observed was four.

The genetic variations observed in the parasite populations in India may provide useful baseline data for future studies on this promising drug resistance locus. Studies using neutral microsatellite polymorphisms flanking the Pfnhe-1 gene along with other loci in the genome are required to understand the basis of this genetic polymorphism whether it is due to genetic drift or selective pressure.

Nucleotide sequence accession numbers. The nucleotide sequences determined in this study have been assigned accession numbers DQ864466 to DQ864486, EF123065, EF123066, and EF442125 to EF442128.