Sequence diversity and natural selection at domain I of the apical membrane antigen 1 among Indian Plasmodium falciparum populations

Abstract

Background: The Plasmodium falciparum apical membrane antigen 1 (AMA1) is a leading malaria vaccine candidate antigen. The complete AMA1 protein is comprised of three domains where domain I exhibits high sequence polymorphism and is thus named as the hyper-variable region (HVR). The present study describes the extent of genetic polymorphism and natural selection at domain I of the ama1 gene among Indian P. falciparum isolates.

Methods: The part of the ama1 gene covering domain I was PCR amplified and sequenced from 157 P. falciparum isolates collected from five different geographical regions of India. Statistical and phylogenetic analyses of the sequences were done using DnaSP ver. 4. 10. 9 and MEGA version 3.0 packages.

Results: A total of 57 AMA1 haplotypes were observed among 157 isolates sequenced. Forty-six of these 57 haplotypes are being reported here for the first time. The parasites collected from the high malaria transmission areas (Assam, Orissa, and Andaman and Nicobar Islands) showed more haplotypes (H) and nucleotide diversity pi as compared to low malaria transmission areas (Uttar Pradesh and Goa). The comparison of all five Indian P. falciparum subpopulations indicated moderate level of genetic differentiation and limited gene flow (Fixation index ranging from 0.048 to 0.13) between populations. The difference between rates of non-synonymous and synonymous mutations, Tajima's D and McDonald-Kreitman test statistics suggested that the diversity at domain I of the AMA1 antigen is due to positive natural selection. The minimum recombination events were also high indicating the possible role of recombination in generating AMA1 allelic diversity.

Conclusion: The level of genetic diversity and diversifying selection were higher in Assam, Orissa, and Andaman and Nicobar Islands populations as compared to Uttar Pradesh and Goa. The amounts of gene flow among these populations were moderate. The data reported here will be valuable for the development of AMA1-based malaria vaccine.

Figure 1

Sliding window plot of the nucleotide diversity per site (π) comparing the level of genetic diversity among the domain I sequences of the parasite from all five study areas. The π values were calculated on DnaSP with window length 100 bp and step size of 25 bp. In all study areas, the maximum diversity was seen between the nucleotide positions 50 and 200 bps. n; number of P. falciparum isolates.

Figure 2

A neighbor-joining (NJ) tree depicting the relationships between different AMA1 haplotypes observed among Indian P. falciparum populations. The distance matrix was prepared using Kimura 2-parameter evolutionary model. Alignment substitutions were considered for analysis while gaps were ignored. Numbers below the line indicate percentage bootstrap values for 1000 replications. The scale bar represents a genetic distance. The partial AMA1 sequence of P. reichenowi [AJ252087], the closet species to P. falciparum was taken as an out group [35].

Figure 3

The linkage disequilibrium (LD) plot showing non-random association between nucleotide variants at different polymorphic sites. The R^{^2} values are plotted against the nucleotide distances with two-tailed Fisher's exact test of significance using DnaSP. The value of LD index (range from -1 to +1) declines with increasing nucleotide distance, indicating that recombination events are taking place. The analysis was performed considering all polymorphic sites. (A) Assam, (B) Orissa, (C) A & N, (D) Goa, (E) UP, and (F) Total.