Plasmodium falciparum Histidine-Rich Protein 2 and 3 Gene Deletions and Their Implications in Malaria Control

Malaria remains the biggest threat to public health, especially among pregnant women and young children in sub-Saharan Africa. Prompt and accurate diagnosis is critical for effective case management and detection of drug resistance. Conventionally, microscopy and rapid diagnostic tests (RDTs) are the tools of choice for malaria diagnosis. RDTs are simple to use and have been extensively used in the diagnosis of malaria among travelers to malaria-endemic regions, routine case management, and surveillance studies. Most RDTs target the histidine-rich protein (PfHRP) which is exclusively found in Plasmodium falciparum and a metabolic enzyme Plasmodium lactate dehydrogenase (pLDH) which is common among all Plasmodium species. Other RDTs incorporate the enzyme aldolase that is produced by all Plasmodium species. Recently, studies have reported false-negative RDTs primarily due to the deletion of the histidine-rich protein (pfhrp2 and pfhrp3) genes in field isolates of P. falciparum. Herein, we review published literature to establish pfhrp2/pfhrp3 deletions, the extent of these deletions in different geographical regions, and the implication in malaria control. We searched for publications on pfhrp2/pfhrp3 deletions and retrieved all publications that reported on this subject. Overall, 20 publications reported on pfhrp2/pfhrp3 deletions, and most of these studies were done in Central and South America, with very few in Asia and Africa. The few studies in Africa that reported on the occurrence of pfhrp2/pfhrp3 deletions rarely evaluated deletions on the flanking genes. More studies are required to evaluate the existence and extent of these gene deletions, whose presence may lead to delayed or missed treatment. This information will guide appropriate diagnostic approaches in the respective areas.


Introduction
The previous two decade was characterized by remarkable progress in the fight against malaria, with more than seven million lives having been saved since 2001 [1], thanks to the wide-scale deployment of malaria control interventions, including but not limited to accurate diagnosis, effective 269 million from 240 million in 2015 [2]. RDTs are fast, easy to perform and interpret, do not need for electricity or equipment, and this has contributed to their widespread use.
Currently, there are about 60 manufacturers and 200 types of RDTs that have been tested and deployed in field settings [19]. All RDTs detect malaria antigen in blood sample placed on a lateral flow chromatographic membrane containing immobilized antimalaria antibodies. The dye-labeled antibody binds the target parasite antigen, and the resulting complex is subsequently captured by another antibody, which is immobilized at the test line. The color appears at the test line. Excess labeled antibody conjugate continues to flow along the strip and is captured at the control line by a secondary antibody. When a sample without target antigens is used, color appears only on the control line.
Malaria RDTs target three Plasmodium antigens that are detected in the same RDT. The histidine-rich proteins 2 and 3 (Pf HRP2/3) are specific for P. falciparum, while Plasmodium lactate dehydrogenase (pLDH) and Plasmodium aldolase are made by all malaria species, hence the name pan-pLDH and pan-aldolase. Thus, in general, RDTs are sold as combinations of pan-pLDH-Pf HRP2/3 or pan-aldolase-Pf HRP2/3 [5].
The pLDH is a terminal enzyme in the glycolytic pathway of both the sexual and asexual stages of the parasite, and each of the four Plasmodium species has its specific isomer of this enzyme, while aldolase is conserved in all the species and is an example of a pan-malarial antigen target [20]. pLDH catalyzes (using NADH, a reduced form of nicotinamide adenine dinucleotide, as a coenzyme) the conversion of pyruvate into lactate and NAD + , (oxidised form of NAD) which is important for parasite survival in red blood cells [21,22]. Since pLDH and aldolase are only produced in living Plasmodium cells and have a short half-life (2-4 days), the presence of these enzymes is an indication of current infection [23]. RDTs that detect pLDH either detect a pan-pLDH which is common to all human infecting species or specific regions unique to P. falciparum or P. vivax-specific pLDH [24]. Although previous studies showed reduced sensitivity of most pLDH-detecting RDTs at extreme temperatures [25], a study in India demonstrated RDT stability and high sensitivity of up to 98% for P. falciparum [26].
Three histidine-rich proteins (HRPs) have been described, namely, HRP1 or the knob-associated histidine-rich protein (KAHRP), HRP2, and HRP3, also known as small histidine-alanine-rich proteins (SHARP) containing 8%, 35%, and 30% histidine repeats, respectively [27,28]. The pfhrp1 gene is associated with the expression of knob-like protrusions on the surface membrane of infected erythrocytes [29]. These knobs are not seen in erythrocytes infected with the ring stages of the parasites, but they develop with the maturation of the parasites to other developmental stages, such as late trophozoites and schizonts [30]. Studies have revealed that the knobs are the sites of attachment to the venous endothelium and are thus responsible for sequestration of infected erythrocytes to deep vasculature, a phenomenon that is common in cerebral malaria [31].
Pf HRP2 is a water-soluble protein produced by the asexual stages and young gametocytes of P. falciparum and can be detected in serum, plasma, cerebrospinal fluid, and urine of infected patients and in the medium of in vitro cultured parasites [32][33][34]. Secretion and synthesis of Pf HRP2 have been shown to begin with immature rings and increase gradually as the asexual parasites mature such that, more than 90% is secreted during schizogony [35,36], and its presence in plasma has been exploited in the design of RDTs for malaria diagnosis [37]. Unlike Pf HRP1, which is found only in knobby (K+) variants of P. falciparum, Pf HRP2 is found in both the knobby and knobless (K-) strain variants [38].
The pfhrp3 gene codes for a homologous protein histidine-rich protein-3 (Pf HRP3). Both Pf HRP2 and Pf HRP3 are very similar in structure, although Pf HRP3 has less histidine content [27]. The structural similarity between the two genes is responsible for cross-reaction of monoclonal antibodies against Pf HRP2 with those of Pf HRP3. Unlike Pf HRP1 and Pf HRP3, Pf HRP2 is abundantly produced and is continuously expressed throughout the parasite life cycle.
The use of Pf HRP2-based RDTs for malaria diagnosis has emerging limitations due to the existence of parasites with pfhrp2 gene deletions which, depending on the extent of the deletions, has led to false-negative results [39][40][41][42]. Initially, experimental studies demonstrated the deletions of pfhrp2 and pfhrp3 genes in laboratory lines [43,44]. Whole-genome scanning and DNA sequence analysis have demonstrated partial or complete deletions of pfhrp2/pfhrp3 genes and/or the flanking genes [40,45].
The deletions involve breakage and rejoining at the unstable sub-telomeric regions of chromosome 8 and 13 for pfhrp2 and pfhrp3 genes, respectively ( Figure 1). Recently, studies have confirmed the existence of deletions in field isolates, thereby threatening the utility of Pf HRP2-based RDTs for malaria diagnosis. Depending on the extent of the deletions, the expression level of Pf HRP2 and Pf HRP3 proteins could cause false-negative results, leading to delay in therapy, and ultimately contribute to increased morbidity and mortality and promote disease transmission. Parasites with pfhrp2 gene deletion have been shown to be reactive to Pf HRP2-based RDTs due to cross-reactivity with Pf HRP3 [46,47]. Few studies have evaluated the presence and extent of these deletions in Africa. In this review, we evaluate the current literature on pfhrp2 and pfhrp3 gene deletions and their potential impact on disease diagnosis, management, control and the eradication goals.

Systematic Review Protocols
The guidelines and procedures of the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) [48] were followed in the current study ( Figure 1).

Search Strategy and Inclusion/Exclusion Criteria
Published scientific articles were retrieved from PUBMED, MEDLINE, and Science Direct. We searched the scientific literature for articles published in English from January 2010 when the first case of P. falciparum parasites lacking pfhrp2 and/ or pfhrp3 deletions was first reported [40], to December 2019.
Keywords used for the search included: Pf HRP2, Pf HRP3, pfhrp2/pfhrp3 gene deletions, and Pf HRP2-RDTs. To filter out articles that were out of scope, we screened only for publications that were reporting on pfhrp2/pfhrp3 deletions. To be included in the analysis plan, a study had to fulfill stipulations as previously suggested [49], i.e., have one initial evidence (microscopy or RDT or both) of the parasite and include a quality control procedure such as co-amplification of a single copy gene, in addition to amplification of pfhrp2 and/or pfhrp3 genes. This is crucial to allow differentiation between lack of amplification due to gene deletions versus no amplification due to PCR failure.

Statistical Analysis
Data were entered in a Microsoft Excel database (Microsoft, Redmond, WA, USA). The metaanalysis was conducted using the random effects model of analysis since it minimizes heterogeneity of the included studies. All of the analyses were implemented using R-3.6.2 metaprop and forest packages in order to estimate the pulled proportion, to investigate publication and other bias and to summarize information on individual studies, respectively. Descriptive statistics, such as bar charts, were used to summarize the distribution of reported deletions using the Statistical Package for Social Sciences (SPSS), version 25.   Gamboa et al. (2010) reported the first evidence of field isolates lacking pfhrp2 and pfhrp3 genes in Peru and demonstrated that these parasites were widespread across the Peruvian Amazon [40]. A huge proportion of the retrospectively collected samples were found to lack the pfhrp2 (41%), pfhrp3 (70%) or both (21.6%) genes. The loss of these genes was confirmed by inability to detect the proteins in an immunological assay [40]. A similar study by Akinyi et al. (2013) in the same region of Peru demonstrated a 20% to 40% increase in the frequency of malaria parasites lacking pfhrp2 over a seven year period [50].

pfhrp2/pfhrp3 Deletions
Pfhrp2 and pfhrp3 gene deletions have also been reported in P. falciparum isolates from Colombia within the Amazon River region, like in Peru [51,52], where the prevalence of pfhrp2 deletions was lower than that of pfhrp3 (18% and 52%, respectively). This is not surprising considering that the majority of parasites with pfhrp2 deletions were collected at the Peru border, where pfhrp2 deletions had been reported before. A study that evaluated the population structure of Colombian P. falciparum at the Colombian pacific coast revealed four different subpopulations (Col-1, Col-2, Col-3, and Col-4) [67]. Similarly, the study by Solano et al. (2015) showed that the parasite isolates segregated into four clusters according to geography (Cluster 2 consisted 100% of the isolates from the Amazonas department, 59% of the isolates from Nariño department segregated into Cluster 4, 68% of isolates collected from Valle were assigned to Cluster 3, and 75% from Cordoba were assigned to Cluster 1). A total of 67% of the pfhrp2 deleted parasites and 61.5% of the pfhrp2/ pfhrp3 double negatives were found to segregate in Cluster 2, which consisted of samples mostly collected from the Amazonas department that borders Peru and Colombia within the Amazon ecological zone, indicating expansion and a common genetic origin of pfhrp2 and pfhrp3 deleted parasites [51].
Although earlier studies in Honduras only reported pfhrp3 and not pfhrp2 deletions, recent reports have documented extensive deletions of both genes [39,54]. The frequency of parasites lacking pfhrp2 and pfhrp3 in South and Central Americas has increased over time, and there is a need to evaluate the selective pressures that favor the survival of parasites that lack these genes [41].
Studies in India reported similar frequencies of parasites lacking pfhrp2 and pfhrp3 and their corresponding flanking regions [55][56][57]. Figure 3 summarizes the prevalence and distribution of pfhrp2, pfhrp3 and the flanking gene deletions in different continents. As shown in Figure 3, the lowest prevalence of pfhrp2/pfhrp3 and the respective flanking gene deletions was reported in Asia. In one of the studies, deletions of the two genes were observed in symptomatic individuals with high parasitemia [56], contrary to previous observations of pfhrp2 and pfhrp3 deletions in asymptomatic individuals with low parasitemia [60].
In Africa, where malaria is endemic and the use of RDTs is widespread, there is scarce information on pfhrp2/3 deletions, although a few surveillance studies have reported on their existence [60,61,63,68,69].
The first cases of pfhrp2 gene deleted parasites were reported in Mali, where deletions were identified in parasitemic individuals whose blood was unreactive to Pf HRP2-based RDTs [60]. In that study, the participants were asymptomatic and the parasites had low multiplicity of infection, a term used to describe strain diversity in an infected individual. Although parasite density may have influenced the variability observed in Pf HRP2-based RDTs, Wurtz et al. (2013) did not find any difference in parasitemia between groups with pfhrp/pfhrp3 deletions and those without deletions, suggesting that, in addition to parasite density, lack of expression of the Pf HRP2 may greatly influence the performance of RDTs. A recent study in Eritrea reported the emergence of false-negative Pf HRP2-based RDTs that was not related to poor quality of RDTs, storage, handling or operator errors [70]. A subsequent study in the same population revealed that most patients were infected with parasites that were lacking pfhrp2 (62%) and pfhrp3 (82%), and that these parasites may have emerged due to local selective pressure since the parasite clusters obtained were not related to South American strains [66]. In Kenya, one study has reported the presence of pfhrp2 deleted parasites in samples collected from asymptomatic children [12]. Curiously, these pfhrp2 deleted samples were positive by Pf HRP2-based RDTs. Although this phenomenon was not reported in the South and Central America studies, it may be due to close reaction of Pf HRP2 monoclonal antibodies with Pf HRP3. In vitro studies have confirmed the detection of parasites with deleted pfhrp2 on Pf HRP2-based RDTs, but this has not been reported widely in field samples [46]. Because of potential failures of RDTs, studies have recommended the performance of two RDTs in parallel as an initial screening test when evaluating pfhrp2/pfhrp3 deletion [49].
Overall, pfhrp2/pfhrp3 deletions have been widely reported in Central and South America, where most studies reported a higher prevalence of pfhrp3 (Lowest = 7.4% in French Guiana and Highest = 96.2% in Honduras) than of pfhrp2 (Lowest = 4.0% in Bolivia and Highest = 40.6% in Peru). Similar deletions were also observed for the upstream and downstream regions flanking pfhrp2 and pfhrp3 genes (Table 1 and Figure 3). The studies included in the current analysis showed extensive heterogeneity in reporting the prevalence of pfhrp2/pfhrp3 and the flanking genes ( Figure 4). This could partly be attributed to the small sample size in these studies. We did not find any publication that reported a lack of deletion in all the genes of interest.    The results of the meta-analysis with random effects model are presented in Figure 4. Based on the estimates, we noted that the proportion of pfhrp2 exon1-2 among the included studies ranges from a minimum of 0.01 (95% CI: 0.00, 0.04) [53] to a maximum of 0.62 (95% CI: 0.47, 0.75) [66]. Furthermore, the pooled proportion of pfhrp2 exon1-2 was 0.10 (95% CI; 0.06, 0.17) ( Figure 4A). The heterogeneity test showed presence of lack of uniformity, p-value = < 0.001. Similarly, the pooled prevalence of Upstream MALP1.230, Downstream MALP1.228, pfhrp3 exon1-2, Upstream MAL13P1.475, Downstream MAL13P1.485 and Double negative for pfhrp2 and pfhrp3 were 18.00%, 3.00%, 32.00%, 38.00%, 35.00% and 8.00%, respectively ( Figure 4B-G). Cheng et al.(2014) published a set of recommendations for reporting pfhrp2/pfhrp3 deletions [49]. These recommendations include an initial screening whereby the sample is confirmed positive for P. falciparum by two expert microscopists and PCR. This is followed by a confirmatory evidence of deletion through PCR amplification of exon 2 of pfhrp2 and pfhrp3 and the region across exon 1 and 2 of both genes.
For quality control, it is also recommended to include PCR amplification of single copy genes such as msp1 and msp2 to rule out negative results due to low-quality DNA or PCR failure. We used these recommendations to evaluate which studies conformed to this protocol. All studies conducted in South America, Central America, and Asia conformed to these recommendations for reporting pfhrp2/pfhrp3 deletions including PCR analysis of the flanking genes, performing RDTs on the suspected samples, and functional analysis through Pf HRP2 ELISA. These recommendations are also in agreement with the recently published WHO guidelines on reporting pfhrp2/pfhrp3 deletions [71] and as such, should be adhered to for the validity of any study on pfhrp2/pfhrp3 gene deletions.
Unfortunately, of the eight studies conducted in Africa, only two followed these guidelines. Most reported on the prevalence of pfhrp2 (Lowest = 1.5% in Mozambique and Highest = 62.0% in Eritrea) and pfhrp3 (Lowest = 1.1% in Kenya and Highest = 82.0% in Eritrea [12,[62][63][64]. The two studies that included analysis of flanking genes in the detection of pfhrp2/pfhrp3 deletion were from Eritrea. However, the study by Menegon et al. (2017) did not include RDT results [65].
Although the upstream and downstream flanking genes for pfhrp2 and pfhrp3 are not well characterized, they have been observed [49], and as such, it is recommended to include these genes when evaluating pfhrp2 and pfhrp3 gene deletions. As shown in Table 1, most studies did not conform to the WHO recommendation for accurate reporting of pfhrp2/pfhrp3 deletions. This may affect the validity of the claims made [49,71].
Apart from gene deletions, sequencing of the pfhrp2 and pf hrp3 genes has revealed extensive variability and different haplotype profiles in parasites from different geographical areas, and this may influence the performance of RDTs [72,73]. For example, the sequencing of these genes has revealed an extensive variation in the number of histidine repeats, copy numbers, and single nucleotide polymorphisms [58,74,75]. The functional utility of these variations needs further investigations, since this variability has not been shown to affect the amount of Pf HRP2 in plasma, which is the basis for parasite detection on RDTs [74,76].

Implication of pfhrp2/pfhrp3 Deletions on RDT Use
The extensive decline in malaria witnessed over the last two decades is attributed to improved diagnosis, effective treatment, and monitoring. The details of the decline are captured in the WHO Global Initiative for malaria control referred to as T3 (Test, Treat, and Track) which provided a framework for malaria control and elimination in malaria-endemic countries [77]. This initiative has contributed to the increased use of RDTs for diagnostic testing, especially in areas where microscopy is not available.
Recent evidence of false-negative RDTs due to the emergence of parasites lacking pfhrp2 and/or pfhrp3 may negatively impact the success of RDTs in malaria diagnosis and could present a major threat to the progress made in the fight against malaria [40,49], and may delay eventual malaria eradication goals. Although the selective pressure-favoring survival of pfhrp2 depleted parasites is not clearly defined, mathematical modeling has suggested that the use of Pf HRP2-based RDTs over the last decade favored the survival of parasites lacking pfhrp2 genes. The model also suggests that selection pressure is high in areas where the prevalence of malaria is low, and the treatment of cases is based on RDT diagnosis [78]. This is because, in low transmission settings, most infections are asymptomatic and are characterized by low-density parasitemia, which is below the detection limit of the RDTs. This means that these infections are not treated and thereby enhance parasite survival [79,80]. Because of the structural homology of Pf HRP2 and Pf HRP3 antigens and the observed cross-reaction of monoclonal antibodies used in the RDTs, it would be expected that Pf HRP3 would compensate for the deletion of pfhrp2. However, because Pf HRP3 is not abundantly produced, this is likely to be beneficial only at high parasitemia [46].
The decline in malaria transmission that has been witnessed in most areas means that regions previously considered endemic may not experience as high a number of malaria cases as before, and most infections will likely be characterized by low-density parasitemia. If the model proposed by Watson is correct, this will increase the number of missed diagnosis by both microscopy and RDT; therefore, individuals harboring pfhrp2 depleted parasites are likely to go undetected [78]. Consequently, this will likely amplify the population of malaria parasites with the these deletions [49]. In peripheral health care facilities where microscopy is not available or among travelers to malaria-endemic areas, a common practice is to rule out malaria in patients that test negative with an RDT. This practice could also aid in the selection of pfhrp2 deleted malaria parasites.
This review demonstrates the emerging threat of pfhrp2/pfhrp3 deleted parasites mostly in areas that have had a history of P. falciparum malaria. This could become a big challenge, especially with the increased use of Pf HRP2-based RDTs. In sub-Saharan Africa, where malaria is still endemic, the extent of pfhrp2/pfhrp3 gene deletions is underreported [71,81]. There is, therefore, a need for more extensive studies to evaluate the true prevalence of pfhrp2/pfhrp3 deleted parasites and the impact they have on malaria diagnosis.