Plasmodium falciparum histidine rich protein 2 (pfhrp2): an additional genetic marker suitable for anti-malarial drug efficacy trials

Background Genotyping of the three Plasmodium falciparum polymorphic genes, msp1, msp2 and glurp, has been adopted as a standard strategy to distinguish recrudescence from new infection in drug efficacy clinical trials. However, the suitability of a particular gene is compromised in areas where its allelic variants distribution is significantly skewed, a phenomenon that might occur in isolated parasite populations or in areas of very low transmission. Moreover, observation of amplification bias has diminished the value of glurp as a marker. Methods The suitability of the polymorphic P. falciparum histidine-rich protein 2 (pfhrp2) gene was assessed to serve as an alternative marker using a PCR-sequencing or a PCR–RFLP protocol for genotyping of samples in drug efficacy clinical trials. The value of pfhrp2 was validated by side-by-side analyses of 5 admission-recrudescence sample pairs from Yemeni malaria patients. Results The outcome of the single pfhrp2 gene discrimination analysis has been found consistent with msp1, msp2 and glurp pool genotyping analysis for the differentiation of recrudescence from new infection. Conclusion The findings suggest that under the appropriate circumstances, pfhrp2 can serve as an additional molecular marker for monitoring anti-malarials efficacy. However, its use is restricted to endemic areas where only a minority of P. falciparum parasites lack the pfhrp2 gene. Supplementary Information The online version contains supplementary material available at 10.1186/s12936-021-04014-4.


Background
Malaria drug efficacy clinical trials are conducted to monitor the continued efficacy of front-line anti-malarial drugs, particularly against falciparum malaria. After administration of full-dose malaria treatment, in vivo drug efficacy is assessed through clinical and parasitological follow-up according to standard protocols [1]. The erstwhile recommended follow-up duration of 14 days was revised to a standard 28 days that can be extended to 42 days or to 63 days for the drugs with a long half-life [1,2]. Given the short incubation period of P. falciparum, longer follow-up periods increase the likelihood of acquiring new infections, especially in areas of high malaria transmission, which can lead to an underestimate of drug efficacy through mistaken classification of newly acquired infections as drug failures. In order to counter this, molecular genotyping strategies based on polymerase chain reaction (PCR) amplification have been developed to correct estimates of drug efficacy in clinical trials by enabling a distinction between recrudescences versus new infections during the follow-up [3]. The genetic markers recommended by the World Health Organization (WHO) for PCR correction are the polymorphic regions of three P. falciparum genes: merozoite surface protein 1 (msp1, block2), merozoite surface protein 2 (msp2, block3), and the repeat region 2 of the glutamaterich protein (glurp) [4]. The amplified allelic variants are classified according to size, and also to allelic type for the msp1 and msp2 [4].
The suitability of each marker depends on the local parasite population to comprise numerous allelic variants with an unbiased distribution. This is likely in areas of high transmission intensity where parasites are genetically diverse and infections are often polyclonal, with a multiplicity of infection (MOI) often reaching 5, but less so in areas of low transmission with low MOI (1)(2), where re-infections by the same genotype might occur more frequently. Furthermore, recent investigations showed that glurp genotyping is affected by amplification bias, thereby diminishing confidence in this marker for PCR correction strategies [5,6]. Reliance on two genetic markers, as compared to three, will diminish the accuracy of PCR correction, and might lead to an overestimation of drug failures, which could prompt an unnecessary change of the first-line treatment, that is advocated when PCR-corrected failure rates exceed 10% [1,6,7].
Defined polymorphic microsatellite markers, that vary by the number of repeated short nucleotide sequences, have been used as an alternative or as an additional standard genotyping marker [8][9][10][11]. Microsatellite markers are usually located in non-coding regions and are considered to be selectively neutral. However, they are known for their instability, with the frequency of repeats within some loci capable of changing during the infection [2,10], which might lead to underestimate drug failure rates [2,10,12]. Moreover, microsatellite markers require a range of 6-13 PCR reactions, based on the number of loci included in the analysis [8,13,14] (Table 1).
The suitability of another polymorphic gene, coding for the P. falciparum histidine-rich protein 2 (pfhrp2), was assessed as an additional marker for the discrimination between recrudescent and new infections. Two genotyping protocols were designed and evaluated, PCR amplification followed by sequencing, and PCR followed by restriction enzyme length polymorphism (PCR-RELP) analysis.

Study setting
In an in vivo study that aimed to assess the efficacy of artemisinin + sulfadoxine-pyrimethamine (AS-SP) drug combination against uncomplicated falciparum malaria in the Tehama region of Yemen [15], recurrent parasitaemia during follow-up was detected in five cases. Of these cases, three were classified as true recrudescence as they were found to carry the same malaria genotypes using the msp1, msp2 and glurp markers. During this study, the impact of pfhrp2 gene variation on the performance of hrp2-based RDT was evaluated in this area. Sequencing analysis revealed high genetic diversity for pfhrp2 with notably low allelic variant frequency distribution [16].

Malaria isolates
Five archived filter paper blood spot admission (prior to treatment baseline, or day 0) and recurrence (day X after ACT administration) sample pairs of P. falciparum infection were included in this analysis. During the 28 days of follow-up: one case recurred at day 14, two at day 21 and two others at day 28.

DNA extraction
Two to three discs (6 mm diameter) of the collected filter paper blood spot were utilized for genomic DNA extraction using Qiagen blood and tissue kit (QIAGEN, DNeasy ® Blood & Tissue Kit, Germany). The DNA was eluted in 35 µL AE elution buffer (10 mM Tris-Cl; 0.5 mM EDTA; pH 9.0) and stored at −20 °C until use.

PCR amplification and sequencing analysis of pfhrp2
Amplification of pfhrp2 was carried out in a singlerun PCR using a specifically designed oligonucleotide primer pair (PfHRP2-F: 5′-TGT GTA GCA AAA ATG CAA AAGG-3′ and PfHRP2-R: 5′ TTA ATG GCG TAG GCA ATG TG-3′) flanking the polymorphic exon-2 region of the pfhrp2 gene, as previously described [16]. Briefly, 3 µL of the genomic DNA was added to total up 50 µL reaction volume containing 20 µL ExPrime Taq Premix (Genet Bio, Korea) and 2 µL of each forward and reverse primers (10 µM). Thermal cycling condition was initiated with a denaturation cycle at 95 °C for 5 min followed by 40 cycles (denaturation step at 95 °C for 30 s, annealing at 57 °C for 40 s and an extension step at 72 °C for 90 s) and a final extension cycle at 72 °C for 10 min. All amplification reactions were performed using MyCycler ™ or T100 Thermal Cyclers (BioRad, Hercules, USA). The PCR products were visualized in 2% agarose gel pre-stained with Sybr ® safe DNA gel stain (Invitrogen, USA) using a Amplicon lengths in base pair were reported against 100 DNA ladder [16]. The PCR product bands were cut out of the gels, purified, and sequenced in both directions. The corresponding consensus amino acid sequences were aligned using BioEdit Sequence Alignment Editor Software (version 7.1.9) and Molecular Evolutionary Genetics Analysis (Mega) software (version 7.0.26) and analysed for the type and number of amino acid repeat units [16]. The sequences for the sample pairs (day 0 sample and day X) from this study were aligned and analysed for similarity in the type and number of the amino acid repeat units. A day X sample was classified as true recrudescent (treatment failure) if it showed 100% similarity in the amino acid repeat type and number as compared to that from the day 0 baseline sample. A difference in the type and/ or in the number of one or more of the amino acid repeat units led the sample to be classed as a new infection.

PCR-RFLP analysis of pfhrp2
Given that sequencing might not be readily available in some settings, a genotyping protocol based on restriction enzyme length polymorphism (RFLP) analysis of the amplified fragment was designed and evaluated. Each PCR product was subjected to digestion using one of two restriction enzymes: Alu I (cutting site AGCT) and Pvu II (cutting site CAG CTG ). Post-digestion products were then analysed in 2.5% agarose gels and the number and size of fragments reported for each sample. Identical digestion patterns between day 0 and day X paired samples indicated a treatment failure, while a difference indicated a new infection.

msp1, msp2 and glurp genotyping
Amplification and analyses of msp1, msp2 and glurp genes were performed using nested PCR according to previously described standard protocols [3,4,17]. Amplicons were visualized in 2.5% agarose gel stained with Sybr ® safe DNA gel stain (Invitrogen, USA) using UV documenting system (Bio-Rad, Hercules, CA, USA). The size of the PCR products for each marker was reported and then grouped into different size bins of 25 bp for msp1 and msp2, and 50 bp for glurp. For the recurrent parasitaemia during the in vivo follow-up period, samples of both day 0 (i.e., pre-treatment) and their corresponding samples collected at day X (i.e. post-treatment) were genotyped and compared for alleles similarities.

Determining allelic frequency distribution for pfhrp2
The type and frequency of the pfhrp2 alleles in the study area were derived using the pfhrp2 sequence of 180 samples obtained from the previous study [16]. A difference of one amino acid or more was considered sufficient to define a new allelic type, and allelic frequencies were then calculated.

Determining allelic frequency distributions for msp1, msp2 and glurp
At least fifty samples are recommended to estimate the allelic frequency and the multiplicity of infection (MOI) of the msp1, msp2 allelic families and glurp genotypes [1]. Thus, fifty archived blood samples of P. falciparum mono-infection were randomly selected and genotyped as previously described [17]. A representative size fragment for each msp1 allelic family (K1, MAD 20 and RO33) and msp2 allelic family (3D7/IC and FC27), as well as glurp allelic variants, were subjected to sequencing to confirm the nature of the allelic variants.

Distribution of msp1, msp2 and glurp alleles
Out of the fifty randomly selected samples, 46 were successfully amplified for all three markers: msp1, msp2 and glurp. Six allelic variants were observed for msp1, five for the msp2 and five for glurp. The frequencies of the 200-225 bp MAD20, and 150 bp RO33 msp1 allelic variants, and the 350-400 bp FC27 and 600-650 bp 3D7/ IC msp2 allelic variants were biased as they were found in 32.5%, 25%, 28% and 39.7% of the examined samples, respectively. Similarly, the glurp alleles 900-1000 and 1000-1100 bp allelic variants were predominant with frequencies of 32.6%, and 30.5%, respectively. Therefore, distinguishing recrudescence from new infection in this area using these alleles might be challenging. Nonetheless, the other 10 allelic variants of msp1, msp2 and glurp had low frequencies (3.8-15%) (Fig. 1). The MOI of the P. falciparum isolates analysed was 1.7, 1.5 and 1.0 for msp1, msp2 and glurp, respectively. Thus, the validity of these three markers for PCR correction was sub-optimal, and likely to over-estimate treatment failures.

Diversity and frequency distribution of the pfhrp2 alleles
In a previous survey of pfhrp2 gene variation [16] conducted in the same area, all isolates (189) were successfully amplified (100% amplification rate) and the pfhrp2 fragment obtained successfully sequenced. The isolates were all found to carry a single allelic variant (i.e., MOI = 1).
Analyses of these pfhrp2 sequences showed a high degree of genetic diversity with 46 distinct allelic variants (477-969 bp, corresponding to 159-323 amino acids) found circulating in the area. Of these, 45 alleles (97.8%) occurred in very low frequency (< 5%), though  (Fig. 2). Details of the hrp2 allelic variants classification are provided in Additional file 1.

PCR correction of recurrent cases based on pfhrp2 as a genotyping marker
Analysis of the paired samples from the five cases with recurrent parasitaemia was conducted by pfhrp2 amplification followed by sequencing. Comparison of the amino acid type repeats and their frequency in the baseline (day 0) and day X sample pairs, indicated that the outcome for two of the patients could be classed as a re-infection (case 2 and case 5), while the other three cases could be classed as recrudescences, i.e., treatment failures (Fig. 3). These results were entirely consistent with the PCR correction that was carried out previously using the standard msp1, msp2 and glurp markers ( Table 2) [15].
Two restriction enzymes, Alu I or Pvu II, have been selected for RFLP analysis, and the digests of the amplified pfhrp2 fragments from the paired samples from the five cases were compared in order to reveal potential differences between allelic variants without the need for sequencing. The presence of an identical number and size of the resulting bands derived from the day 0 and day X sample pairs indicated a recrudescence (cases 1, 3 and 4), while distinct patterns for the paired samples indicated a re-infection (cases 2 and 5) (Fig. 4A  and B). These results were consistent with those based on genotyping based on the analysis of the pfhrp2 repeats sequences as well as that based on the standard msp1, msp2 and glurp genotyping protocols.

Discussion
Malaria drug efficacy clinical trial in endemic areas remains the gold standard method for monitoring changes in the efficacy of antimalarial drugs, and their outcomes guide the treatment policy adopted nationally. Correction of these outcomes via parasites genotyping, in which treatment failures are distinguished from re-infections that might have arisen during the follow-up period, has substantially improved outcome accuracy. In recent years, three genetic markers with standardised genotyping protocols have been adopted [1] in order to ensure consistency and comparability between studies. The validity of one of these markers (glurp) has been recently put into question [5], and in some regions low allelic diversity for the one or other of the other two markers (msp1 and msp2) diminishes their adequacy [18][19][20]. The likelihood of low genetic diversity increases in areas of low transmission intensity or in those where anti-malarial measures are leading to near elimination. In such cases, additional polymorphic genetic markers will need to be considered to achieve PCR correction of trial outcomes.
Given the very high diversity observed for the variable repeat region of the pfhrp2 exon in many malaria endemic areas [21,22], including those with low transmission intensity, its usefulness was assessed as a potential additional genetic marker for PCR correction analyses. Although amplicon size could be sufficient to discriminate between two isolates, two post-amplification strategies with a higher resolution were employed to distinguish between the various allelic variants: amplicon sequencing, an onerous though highly accurate method, and RFLP analysis, a less sensitive but more field-friendly approach. In the present study, the potential of pfhrp2 as an  additional marker for PCR correction analyses in Yemen has been demonstrated. Genotyping of the paired samples from the five cases with recurrent episodes confirmed the classification (recrudescence or re-infection) that was obtained using the three standard markers. Moreover, in this case, this was equally achieved by direct sequencing of the PCR product as well as through restriction enzymes digestion of the amplicons. It must be noted that this preliminary study concerned only 5 cases that were sampled about seven years ago. Therefore, validation of this marker will require further studies to confirm its usefulness. Nevertheless, it must be stressed that the value of pfhrp2 as a genotyping marker is severely compromised in areas where this gene is fully or partially deleted in a large proportion of the circulating parasites. Plasmodium falciparum parasites lacking pfhrp2 were first noted in Peru, where 41% of the samples collected proved negative for this gene [23]. Similar observations were made in the following year with samples collected in various South American countries [24][25][26][27][28][29], though much lower proportions of such parasites were recorded in other areas at that time [30][31][32]. A recent systematic review on the prevalence of pfhrp2 negative parasites indicated that their prevalence is highly variable and potentially equally focal [33]. Thus, pfhrp2 must not be used as a maker for PCR correction in areas where the prevalence of the parasites lacking the exon 2 of this gene exceeds a few percentage points (10% as a conservative suggestion). Nonetheless, given its high genetic diversity with allelic variants that can be distinguished by amplicon length, restriction digestions patterns and eventually sequencing, pfhrp2 can serve as a first marker to be analysed from paired samples; any cases thereby clearly classed as re-infection could then be excluded from those that would require genotyping by the other markers (msp1 and msp2), thus saving time and resources.

Conclusion
The polymorphic pfhrp2 gene has the potential to act as an additional genetic marker for inclusion in the PCR genotyping protocols employed to correct the outcomes of in vivo efficacy drug trials conduction in malaria endemic areas. However, its use is restricted to endemic areas where only a very small minority of P. falciparum population lack the pfhrp2 gene.