Sixty-eight P. falciparum samples that were collected during a previous study conducted between September 2008 and September 2009 in the city of Puerto Lempira, which is located in the province of Gracias a Dios, Honduras, were available for this retrospective investigation [11]. In the original study, filter paper blood spots were collected from febrile patients between the ages of six months and 60 years who had uncomplicated P. falciparum mono-infection and had provided written informed consent at the time of enrollment. These patients participated in a clinical trial to assess the efficacy of chloroquine. Further information on the patients and study has been published previously [11]. The study was approved by the institutional ethical review committee of the Ethics Committee of the Medical Sciences Faculty of the National Autonomous University of Honduras (UNAH-IRB 00003070). Investigators from the Centers for Disease Control and Prevention obtained institutional permission to use these samples for the current study under a non-research determination.

Genomic DNA was extracted from dried blood spots on filter paper using the Qiagen™ kit (QIAGEN, Valencia CA) according to the manufacturer’s instructions and resuspended in 150 μl elution buffer.

In order to determine if there were sufficient quantities and quality of DNA present in the samples collected, two genes were amplified, 18s ribosomal RNA (18S rRNA) and merozoite surface protein 2 (msp2), using established PCR protocols [12]. An assumption was made that the successful amplification of both genes was indicative of a reasonable quantity and quality of DNA that would allow for the amplification of pfhrp2, pfhrp3 and their respective neighbouring genes. Therefore, only those samples that successfully amplified for these two genes were included for final analysis of pfhrp2 and pfhrp3 gene deletion. The PCR amplification of the 18S ribosomal RNA gene was performed using methods previously described [12]. Primer sequences for this nested reaction were as follows: 5′CCT GTT GTT GCC TTA AAC TTC3′ and 5′TTA AAA TTG TTG CAG TTA AAA CG3′ for the primary reaction and 5′ TTA AAC TGG TTT GGG AAA ACC AA ATA TAT T 3′ and 5′ ACA CAA TGA ACT CAA TCA TGA CTA CCC GTC 3′ for the secondary reaction. Briefly, PCR reactions were performed in 25 μl total volume containing 10× buffer, 4 mM MgCl2, 200 μM dNTPs, 250 nM primers, 1.25 units of Taq Polymerase (New England Biolabs, Ipswich, MA, USA), and 1–3 μl of DNA template. After confirmation of Plasmodium genus, species-specific primers were used to confirm P. falciparum infection.

The msp2 gene was amplified using a previously described method [13]. Briefly, a nested PCR reaction was used in which primary amplification targeted a conserved region of this gene and a secondary PCR step that amplified a polymorphic region of msp2 (3D7 and FC27 families). The primer sequences for the primary reaction were 5′GAA GGT AAT TAA AAC ATT GTC 3′ and 5′GAT GTT GCT GCT CCA CAG3′ and 5′GAG TAT AAG GAG AAG TAT G3′ and 5′CTA GAA CCA TGA ATA TGT CC3′ for the secondary reaction. Confirmatory PCR reactions for msp2 were performed in 20 μl total volume containing 10× buffer with MgCl2, 200 μM dNTPs, 250 nM primers, 1.25 units of Taq Polymerase (New England Biolabs, Ipswich, MA, USA), and 1–3 μl of DNA template. When PCR amplification was not successful for either gene, it was repeated to confirm the result. However, if the first two PCR amplification results were discordant, the amplification was performed a third time and the two concordant results were scored as the final result.

Nested PCR amplification of each of the fragments spanning exon 1, the intron, and exon 2 of pfhrp2 and pfhrp3 (Figure 1) were performed using the primers and reaction conditions described in Table 1. The primers listed in Table 1 were designed around highly conserved regions flanking each of the genes of interest in order to minimize the possibility of variation in amplification due to gene polymorphisms. Genes immediately flanking pfhrp2 upstream (MAL7P1.230) and downstream (MAL7P1.228), and those flanking pfhrp3 upstream (MAL13P1.475) and downstream (MAL13P1.485), were also amplified using the primers and reaction conditions described in Table 1.Figure 1

Schematic of (A) Pfhrp2 and (B) Pfhrp3 and their respective flanking genes. In PlasmoDB version 8.0, from which gene information was obtained, pfhrp2 was located on chromosome 7 although in a recent update of PlasmoDB this gene and its flanking genes were reassigned to Chromosome 8. Pfhrp3 was reported to be immediately flanked by the indicated genes, all of which are on Chromosome 13. Arrows indicate the gene fragments that were amplified.

The prevalence of pfhrp2, pfhrp3, and flanking gene deletions was determined by dividing the number of isolates that had deleted the gene by the total number of isolates determined to be positive for both 18S rRNA and msp2.

To determine the population structure of P. falciparum parasite samples, we genotyped samples using seven neutral microsatellite markers used in previous studies [15-18]. Since only a limited amount of DNA was available from the samples, we performed whole-genome amplification of the samples using the Repli-G amplification kit (Qiagen, Valencia CA). The whole genome amplified samples were used for the amplification of the following neutral microsatellite loci: TA1 and TA109, both of which are located on chromosome 6; poly α (chromosome 4); PfPK2 (chromosome 12) and 2490 (chromosome 10); C2M34 (chromosome 2) and C3M69 (chromosome 3) loci. The amplification products were labeled with fluorescent dyes (FAM or HEX) and their sizes assayed on an Applied Biosystems 3130 xl sequencer. The fragments were then scored using GeneMapper software v.3.7 (Applied Biosystems, Foster City CA) with default microsatellite settings, where bands smaller than 500 relative fluorescence units (rfu) were defined as background. Samples for which we obtained no amplification in some loci were re-analysed to complete the microsatellite haplotype profiles. If a locus did not amplify after two rounds of PCR, the result was recorded as negative.

Overall parasite genetic diversity was examined by using neutral microsatellite data to calculate the microsatellite locus-specific heterozygosity and number of alleles per locus (A) in the Excel Microsatellite Toolkit. Any microsatellite locus that did not amplify was assigned a null value (350) for the purpose of these calculations. To determine locus-specific heterozygosity, we used the virtual heterozygosity estimate (H_E), defined as H_E = [n/(n − 1)][1 − Σp_i²], where n is the number of isolates analysed and p_i is the frequency of the i-th allele in the population. H_E gives the average probability that a pair of alleles randomly obtained from the population is different and the values range between 0 and 1.

To determine the population structure of P. falciparum isolates collected in Puerto Lempira, a Bayesian approach was used to infer the number of genetically related clusters (K) from the individual microsatellite profiles generated with the seven neutral microsatellites. Only samples that were determined to be singly infected (based on obtaining a single peak for each locus by neutral microsatellite analysis) were used for cluster analysis (N = 65). The calculation was performed using Structure v2.3.3 [19] in which the likelihood of finding between one and ten clusters in this population (K = 1 to K = 10) was tested. Twenty replicates of the clustering algorithm for each value of K were performed with a burn-in period of 10,000 iterations and 100,000 Markov Chain Monte Carlo replications, using the admixture model with correlated allele frequencies [20]. The most likely number of clusters was defined by calculating the ΔK value as described by Evanno et al. [21] and results from Structure were entered into the Structure Harvester program [22].

Population pairwise F_ST calculations were performed in Arlequin v3.11 [23]. The exact test of population differentiation was used with 1000 permutations and a significance level of 0.05.

All 68 P. falciparum parasite isolates from Honduras were initially tested for the presence of 18S rRNA and msp2 and showed positive amplification, confirming the presence of good quality DNA. Pfhrp2 and its flanking genes, MAL7P1.230 and MAL7P1.228, were found to be present in all parasite isolates.

Thirty out of 68 isolates (44.1%) tested negative for the pfhrp3 gene while 32 isolates (47.1%) showed deletion of the Mal13P1.475 gene that is located upstream of pfhrp3 (Figure 2). Thirteen isolates had deleted the Mal13 P1.485 gene, located downstream of pfhrp3 (Figure 2).Figure 2

Prevalence of deletions in pfhrp3 and neighboring genes in P. falciparum isolates from Puerto Lempira, Nicaragua. The map shows the location of Honduras in relation to neighboring countries in Central America. All parasite isolates analysed were found to be positive for pfhrp2 and its flanking genes (not shown). The three pie charts shown illustrate the proportion of parasite isolates with deletions in pfhrp3 and its neighboring genes. The percentages shown represent proportions of samples out of the total samples that were 18S rRNA- and msp2-positive.

We performed neutral microsatellite genotyping and cluster analysis to determine if the population structure correlated with pfhrp3-deleted parasites. Structure analysis predicted the presence of at least two clusters of parasites (K = 2; Figure 3). Twenty-seven samples belonged to the first cluster (designated Cluster 1) while 37 were assigned to Cluster 2 (Figure 3). One isolate belonging to Cluster 1 was MAL13P1.475-negative (3.7%), six were MAL13P1.475/pfhrp3-negative (22.2%) and one had deleted all three genes (3.7%) (Table 2). Among the samples assigned to Cluster 2, two were MAL13P1.475-negative (5.3%), twelve isolates were MAL13P1.475/pfhrp3-negative (32.4%) while eight were MAL13P1.475/pfhrp3/MAL13P1.485 triple-negative (21.6%) (Table 2). In order to assess genetic differentiation due to population substructure, pair-wise F_st values were calculated, and found that Clusters 1 and 2 presented moderate genetic differentiation (0.12146; p-value < 0.05).Figure 3

Bayesian cluster analysis of P. falciparum singly-infected samples collected from Puerto Lempira, Honduras (N = 65). The predicted number of likely clusters (K) for the samples was K = 2. Each color corresponds to a population cluster as classified by Structure v2.3.3, and each individual isolate is represented by a vertical bar. The Y axis represents the estimated proportion of membership of an individual to each population.

H_E, a measure of genetic variation at each locus, was between 0.2 and 0.7 with relatively few allelic forms of each marker (A) identified (Table 3). This is indicative of low genetic diversity in this parasite population. The number of alleles per locus in this population ranged from three to five (Table 4).Table 3

Microsatellite diversity for seven P. falciparum neutral microsatellite markers