Background
Yunnan was one of the only two provinces in China where malaria was indigenous endemic cases as of 2013 [
1,
2]. The number of malaria cases diagnosed by parasitology method (including microscopy and genetic test) in Yunnan Province took over 17.3% (533/3078) of China, and 45.0% (9/20) of highly endemic counties in 2014 [
2]. Although the number of reported malaria cases in the Yunnan Province has been the highest in China since 2016, but all the cases were imported [
3]. Among them, Myanmar (Burma) was the largest source of imported malaria cases, taking over 84.9% (477/562), while falciparum malaria was the main type of imported malaria from Africa (92.9%, 79/85). The patients were mainly farmers and businessmen who work abroad. Cases were found mainly in Tengchong, Ruili and Yinjiang counties along the China-Myanmar border. Therefore, Yunnan Province pays special attention to timely diagnosis and discovery of malaria spreading sources with measures such as vector control for stop malaria spreading and sensitivity monitoring for anti-malaria drugs [
4‐
7]. A specific strategy, focused on the primary medical health institutions and individual clinics across the province, is to promote the use of immune rapid diagnostic tests (RDTs), such as Wonfo rapid diagnostic cassette which sensitivity was almost 95% for
Plasmodium falciparum in 2008 [
8]. The use of RDTs are up to ten thousands cassettes per year in order to ensure that the malaria cases in the far rural of Yunnan Province get timely screening, but malaria diagnosis must be confirmed pass through various stages of county-level, prefecture-level, and province-level malaria diagnostic laboratory. The confirmation methods include further parasitological microscopy by professionals in combination with genetic testing, in order to reach an accurate diagnosis of malaria and a general quality evaluation of RDTs diagnosis [
6,
7,
9]. After several years of observation in Yunnan Province, false negative and false positive malaria cases by RDTs diagnosis occurred frequently, compared with the gold standard microscopic examination method. A systematic evaluation of the accuracy of RDTs diagnosis of malaria in Yunnan Province showed that nearly half of the RDTs were insensitive to about 3–33% of
P. falciparum infections despite the tendency to overestimate the sensitivity and specificity [
10]. The World Health Organization (WHO), after six rounds of RDT quality assessments, pointed out that the
P. falciparum histidine-rich protein 2 (
PfHRP2) gene deletion may be the mainly biological factor accounting for failed diagnoses of falciparum malaria. Meanwhile, the WHO appeal to the regions with falciparum malaria epidemic worldwide for promptly clarification and validation of the
PfHRP2 (genes) deletion as well as its influence on the accuracy of RDTs diagnosis of malaria [
11].
The target proteins of malaria RDTs diagnosis mainly include
PfHRP2,
P. falciparum lactate dehydrogenase (pLDH) and
P. falciparum aldolase (ALD). The first two are the detection targets of most malaria RDT products [
10,
12], and
PfHRP2 is a specific diagnosis target protein for falciparum malaria. Unfortunately, since the existence of
P. falciparum strains with
pfhrp2 gene deletion was first reported in the 1980s [
13‐
15], the phenomena of
pfhrp2 gene deletion has been found in Peru [
16], Mali [
17], India [
18,
19], Philippines [
20], Senegal [
21], Brazil [
22], Yemen [
23], Honduras [
24]. Amoah et al. found that in Ghana the false negative rate by RDTs diagnosis for falciparum malaria was 52%, and the proportion of undetected
pfhrp2 gene was as high as 40%. In contrast, in areas with a false negative rate of 2%, only 22% of
pfhrp2 genes were undetected [
25]; hence, Amoah et al. speculated that the spread of
P. falciparum without the
pfhrp2 gene had increased the incidence of errors in RDT diagnosis of falciparum malaria in Ghana. The studies conducted by Akinyi et al. and Wurtz et al. indicated these phenomena of a combination of RDT diagnosis failures for falciparum malaria, and the
pfhrp2 gene entire deletion in
P. falciparum isolates from Peru and Senegal [
16,
26]. In Yemen [
23], Angola [
21], and India [
19], entire deletion rates of the
pfhrp2 gene exon2 regions reached 9.5%, 10%, and 23%, respectively, indicating that these countries may have a higher risk of misdiagnosis with the
pfhrp2 (gene)-deficient strains when using
PfHRP2-target protein RDTs for falciparum malaria diagnosis. In 2014, Yang et al. detected the polymorphisms of
pfhrp2 gene in 20
P. falciparum isolates collected from the China-Myanmar border [
27]. In 2015, Dong et al. also found the possibility of partial and entire polymorphisms of
pfhrp2 gene deletion in falciparum malaria case isolates from Yunnan Province [
28]. Certainly, there are exceptions across the world, for example, Kenya has not yet found any entire deletion of
pfhrp2 gene in
P. falciparum [
29,
30]. However, given that Yunnan Province has not explored the reasons for the failure of RDT diagnoses of falciparum malaria, the polymorphism associated with
pfhrp2 gene deletion and its changes need to be further verified. More importantly, in response to the WHO’s appeal for the selection and improvement of RDTs, it is essential to understand the prevalence of
pfhrp2 gene deficiency,
PfHRP2 protein deficiency, the extent of antigenic variation, and to provide evidence for Yunnan Province to improve the quality of falciparum malaria RDT diagnoses [
11,
30]. This study aimed to systematically analyse the data from an ongoing molecular epidemiological investigation on
pfhrp2 gene deletion in Yunnan Province.
Discussion
pfhrp2 gene (1064 bp) is located in the chromosome 8 telomere area [
35], and consists of one non-coding region, one intron area, and one histidine and alanine tandem repeat as dominant in coding regions. The alanine and histidine tandem repeats about 1000 bp, and the secretory area at upstream of the coding area 5’- terminal constitutes the exon2 region (a total of 305 coding amino acids). Due to the instability of the telomeres, the frequent recombination in
pfhrp2 gene region is prone to lead to increased mutations and a more complex diversity of amino acid tandem repeats [
16,
37‐
40]. The deletion of
pfhrp2 gene exon2 [
15] and the protein variation of expressed
PfHRP2 not only affect the accuracy of RDTs diagnosis falciparum malaria against
PfHRP2 target protein, but also lead to the potential harm of changing the habits of
Plasmodium and even changing the prevalence intensity of malaria [
41].
PfHRP2 is not only involved in transforming the toxicity for pigment to poison
Plasmodium parasites, but also plays a role in reshaping the structure of the erythrocyte membrane to enhance the immune escape of malaria parasites [
42‐
44].
In this study, the polymorphism of
pfhrp2 gene in the 57–301 aa coding region, which is also a dense repeat region of alanine and histidine in the exon2 region, of
P. falciparum isolates in Yunnan province in the past 6 years was analyzed. Of the 306 blood samples, 250 samples had obtained the target DNA sequence (345–927 bp), which are similar in length from the geographical isolates in Senegal [
21], Nigeria [
23], Yemen [
29], and Mozambique [
45,
46]. There were 151 haplotypes in the DNA sequence, with a PI of 0.169, which is higher than those in Mali (0.005) and Angola (0.011) [
21], and lower than that in Eritrea (0.580) [
45]. Furthermore, the haplotype which there was only one DNA sequence accounted for 33.2% (83/250) of the 250 DNA sequences, which was less than was observed by Deme et al. (43.3%, 29/67) [
21] and Baker et al. (56.6%, 259/458) [
29] in their analyses of mixed samples in Africa. This result suggests that the differentiation of
pfhrp2 gene in the sample isolates was greater than that previously reported in African isolates [
21,
29], which may be due to the inclusion of complex geographical populations and highly differentiated isolates in the sample. The nucleic acid diversity index (PI) of
pfhrp2 gene from the Yunnan isolates was as high as 0.249, which is the highest among Myanmar, African and Yunnan groups.
The 250
PfHRP2 peptide chains derived from the DNA sequence were between 115 and 309 aa in length, with an average of 245 aa, from total cases isolates. The length of 242 aa from the Myanmar isolates was consistent with the study conducted by Baker et al. [
29], but shorter than Haiti’s 262 aa, Brazil’s 261 aa and Vietnam’s 274 aa. This result further enriched the polymorphism of the
pfhrp2 gene exon2 region in Southeast Asia. It is interesting to note that for the same infection across the same period, when we compared the
pfhrp2 gene exon2 area and DNA sequences, and found that the peptides and amino acids of
pfhrp2 with the same chain length as other infection isolates have different sequence structures. A prompt investigation of
pfhrp2 gene mutation and gene exon2 region diversity may have potential gene traceability between different
P. falciparum isolates.
Regarding the
PfHRP2 peptide amino acid repeat feature recognition, all 250 peptide chains showed 19 kinds of repeats, including type 1, type 2 and its three variation types, type 3 and its one variation type, type 4, type 5, type 6, type 7 and its two variation types, type 8 and its one variation type 10, type 12 and type 13 (Table
1). The repeats of types 9 and type 11 was not observed in this study. These repetitive features differed from those 13 types obtained from India isolates by Kumar et al. [
18,
47], 12 repeating characteristics from Senegal isolates by Deme et al. [
21], 14 repeating patterns from mixed isolates collecting from many countries by Baker et al. [
29], and 39 repeating patterns from Kenya isolates by Nderu et al. [
30]. However, in all these studies, amino acid residues repetition of types 1–8, type 10 and type 12 were found. The pattern starting with type 1 repetition and ending with type 12 repetitions were relatively fixed, and they were found in all peptide chains. However, further comparison with different studies showed that the identified amino acid repeat types of the
PfHRP2 peptide chain must be in accordance with the stable identification criteria, especially when the amino acid repeat types in
PfHRP2 exon2 region were used for predicting RDT reactivity for the
PfHRP2 peptide chain. Although the amino acid repeat sequence of
PfHRP2 peptide chain were summarized as type 1, type 2, type 3, type 4, type 5, type 6, type 7, type 8, type 10, type 12, type 13, type 14 and other variants [
18,
20,
30], the recognition and counting of repeat types can only be done manually. Therefore, potential counting differences might exist caused by different counting strategies. For example, type 2 (AHHAHHAAD) was counted as a linkage between type 4 (AHH) and type 7 (AHHAAD), while type 14 (AHHAHHATD) was counted as a linkage between type 4 (AHH) and type 6 (AHHATD), which resulted in trading off and taking turns of type 2, type 4, type 6, and type 7. Due to the complex recognition principle of “long rather than short” in this study, the differences may occur in the recognition and counting of type 2, type 7 and type 4 repetitions compared with those of Yang et al. [
27] and Li et al. [
48]. For the detection of blood samples from falciparum malaria cases in Yunnan province, the average number of repetitions of type 7 and type 4 in this study were 4.8 times and 1.0 time, respectively, while Yang’s type 7 was 15.6 times, with no type 4 repeat characteristics [
27], showing a significant difference.
Studies have demonstrated that antigen variation is a risk factor for the unstable quality of falciparum malaria diagnosed by RDTs based on immune response [
26,
49], and the epitope region of the
PfHRP2 peptide chain is thought to be mainly confined to its central region [
50]. Therefore, rapid prediction of RDTs testing sensitivity for
PfHRP2 peptide chain by type 2 repeat times and type 7 repeat times have become a trend [
18,
20,
26,
29,
30]. Studies conducted by Kumar et al. [
18] and Wurtz et al. [
26] found that when the products of type 2 and type 7 repeat times are < 43 (group C chain), RDTs will find it difficult to detect the presence of this
PfHRP2 chain. Fortunately, the
PfHRP2 peptide chain belonging to group C was not the dominant species in the current study samples. In this study, the
PfHRP2 peptide chain of group C only accounted for 27.3% (68/250), but it was still higher than that of the samples from Senegal (7.4%, 9/122) [
26] and in Kenya (9.8%, 24/244) [
30]. In contrast, the dominant species were
PfHRP2 peptide chains belonging to group B, accounting for 50.0% (125/250), which was lower than 71.3% (87/122) [
26] and 75.8% (185/244) [
30] obtained from previous studies. This suggested that there may be more RDT detection failures than those for African isolate alone in this study, which was mixed with many Myanmar isolates (59.6%, 149/250). Unfortunately, matched pair study between falciparum malaria diagnosed using RDT in Yunnan province and the polymorphic structure of
PfHRP2 peptide chain have not been implemented. Therefore, it is not possible to infer whether
PfHRP2 peptide chain variation affects falciparum malaria diagnosis in Yunnan province.
In this study,
pfhrp2 gene was not amplified in 15.36% (47/306) of the samples when extracting
P. falciparum genomic DNA was confirmed to be normal from these samples. Still, the author could not ensure that the
P. falciparum isolates in these samples can be considered entire deficiency of
pfhrp2 gene. First,
pfhrp2 gene deletion was not verified through Cheng’s method [
51]. Second, although the nested PCR method was used in this study to amplify
pfhrp2 gene with higher amplification efficiency than Kumar et al. [
18] and Baker et al. [
20], the amplified fragment only covered the region from the intron to 248 bp downstream of
pfhrp2 gene, and the sensitivity to confirm
pfhrp2 (gene) deletion may not have been high enough. Parr et al. [
52] and Trouvay et al. [
53] adopted whole genome sequencing, while Gupta et al. [
46] and Ranadive et al. [
54] amplified the region to form the upstream and downstream of
pfhrp2 gene extension [
51], while the phenomenon of
pfhrp2 gene deletion was not found or rarely found. However, more phenomenon of
pfhrp2 gene deletion were found only in amplifying exon2 region alone [
20,
36,
55], which suggests that analysis of the exon2 region of
pfhrp2 gene alone may be less sensitive than whole-genome or upstream and downstream extension assays for confirmation of
pfhrp2 gene entire deletion. Third, the failure to amplify the
pfhrp2 gene above 47 samples does not rule out the effect of low
Plasmodium density. Gupta et al. indicated that the analysis on
pfhrp2 gene deletion was unreliable when
Plasmodium density is only 3 ~ 330 parasites/µl. Therefore, the study conducted by Beshir et al. excluded samples with
Plasmodium densities < 5 parasites/µl during validation of
pfhrp2 gene deletion. In conclusion, the certification of
pfhrp2 gene entire deletion must be rigorous, as the WHO requires that programme must adjust their RDTs product when the proportion of
pfhrp2 gene entire deletion of
P. falciparum achieve to 5% in a population within an area [
56].
This study comprehensively analysed the
pfhrp2 gene polymorphism from falciparum malaria cases isolates from January 2013 to December 2018 in Yunnan province. The current
pfhrp2 gene polymorphism situation from falciparum malaria case isolates in Yunnan province was reflected due to good continuity and systemic characteristics of these samples. The results of the study will be the basis for further verification of
pfhrp2 gene entire deletion and screening of optimal epitopes of
PfHRP2 peptide chain. However, this study include several limitations: (1) due to the inability to determine the
Plasmodium density in some blood samples, the effect of low protozoan density on the confirmation of
pfhrp2 gene deletion cannot be ruled out; (2) polymorphic and deletion of
pfhrp3 gene, which are considered as early signs of
pfhrp2 gene deletion [
15], were not conducted due to limited human resources; and (3) whether the
PfHRP2 peptide chains exists or not in blood samples was not examined in parallel with RDT products in the study.
The study conducted by Watson et al. indicated that low prevalence of malaria and the high frequency for treatment could increase the risk for inappropriate use of antimalarial drugs because malaria might be misdiagnosed by RDTs, which will also lead to transitionally screening on
pfhrp2-deficient plasmodium [
57]. Currently, there was very low prevalence of malaria and high frequency of malaria-associated medical treatment in China. However, China was not a part of the WHO’s global RDT quality monitoring network, and the systematic quality tracking of RDT products were not conducted. Therefore, whether the use of RDTs will accelerate screening for
pfhrp2-deficient
Plasmodium deserves attention. Further rapid verification of the entire deletion of
pfhrp2 gene from
P. falciparum in China deserve higher priority. Fortunately, a more sensitive and specific sequencing analysis method for
pfhrp2 gene deletion has developed.