DNA recovery from used malaria RDT to detect Plasmodium species and to assess Plasmodium falciparum genetic diversity: a pilot study in Madagascar

Rapid diagnostic tests (RDT) are widely used for malaria diagnosis in Madagascar, where Plasmodium falciparum is the predominant species. Molecular diagnosis is essential for malaria surveillance, but requires additional blood samples for DNA extraction. Used RDTs is an attractive alternative that can be used as a source of DNA. Plasmodium falciparum genetic diversity and multiplicity of infection, usually determined by the genotyping of polymorphic regions of merozoite surface proteins 1 and 2 genes (msp1, msp2), and the repeated region RII of the glutamate-rich protein gene (glurp) have been associated with malaria transmission levels and subsequently with the impact of the deployed control strategies. Thus, the study aims to use RDT as DNA source to detect Plasmodium species, to characterize Plasmodium falciparum genetic diversity and determine the multiplicity of infection. A pilot study was conducted in two sites with different epidemiological patterns: Ankazomborona (low transmission area) and Matanga (high transmission area). On May 2018, used RDT (SD BIOLINE Malaria Ag P.f/Pan, 05FK63) were collected as DNA source. Plasmodium DNA was extracted by simple elution with nuclease free water. Nested-PCR were performed to confirm Plasmodium species and to analyse P. falciparum msp1, msp2 and glurp genes polymorphisms. Amongst the 170 obtained samples (N = 74 from Ankazomborona and N = 96 from Matanga), Plasmodium positivity rate was 23.5% (40/170) [95% CI 17.5–30.8%] by nested-PCR with 92.2% (37/40) positive to P. falciparum, 5% (2/40) to Plasmodium vivax and 2.5% (1/40) to P. falciparum/P. vivax mixed infection. Results showed high polymorphisms in P. falciparum msp1, msp2 and glurp genes. Multiple infection rate was 28.6% [95% CI 12.2–52.3%]. The mean of MOI was 1.79 ± 0.74. This pilot study highlighted that malaria diagnosis and molecular analysis are possible by using used malaria RDT. A large-scale study needs to be conducted to assess more comprehensively malaria parasites transmission levels and provide new data for guiding the implementation of local strategies for malaria control and elimination. Trial registration Retrospectively registered

Background Despite the deployment of considerable efforts and various strategies to control and eliminate malaria, this disease remains one of the main cause of morbidity and mortality worldwide [1]. In 2020, 241 million cases were reported leading to 627,000 deaths [2]. Sub-Saharan Africa, including Madagascar, still remains the most affected region [2]. Among the five Plasmodium species infecting humans, Plasmodium falciparum is the most prevalent malaria species in sub-Saharan Africa and the most virulent species leading severe malaria [3].
Rapid Diagnostic Tests for malaria (RDT) are commonly used for malaria diagnosis in field by community health workers to promptly manage malaria cases. Malaria molecular diagnosis are essential for epidemiological surveillance, but requires to collect additional blood samples for DNA extraction. Used RDTs is an attractive alternative that can be used as a source of DNA [4][5][6].
Plasmodium falciparum genotyping remains an important tool for studying types and numbers of parasite clones present in an infection. Currently, this approach is mainly used to investigate the genetic diversity of parasites in human infections and subsequently estimate the malaria transmission intensity [1,[7][8][9][10]. Moreover, numerous studies have demonstrated an association between high multiplicity of infection (MOI) and malaria severity, especially in high malaria transmission areas [11]. Usually, P. falciparum genetic diversity is determined by the genotyping of the polymorphic regions of the block 2 of merozoite surface protein-1 (msp1), block 3 of merozoite surface protein-2 (msp2) and the RII repeated region of the glutamic rich protein genes (glurp) [12][13][14][15].
In Madagascar, four to eight malaria ecozones are described according to their different epidemiological profile that range from low to high transmission [16,17]. Regular molecular and epidemiological monitoring over time and space of the genetic diversity of P. falciparum populations in association with malaria phenotypes (uncomplicated malaria, severe malaria and asymptomatic malaria cases) is crucial to evaluate the impact of malaria control interventions and to guide the deployment of local tailored strategies for elimination [1,11,18,19].
Hence, this study aims to valorize blood samples collected onto RDT as DNA source for detecting Plasmodium infections, assessing the genetic diversity of P. falciparum populations and estimating the MOI in P. falciparum isolates collected from symptomatic patients seen in health centres in two sites with different epidemiological patterns: Ankazomborona, located in a low transmission area and Matanga in a high transmission area.

Study sites and sample collection
The study was conducted on May 2018 in two sites: in Ankazomborona, district region of Marovoay, located on the west coast of Madagascar (16° 07′ 00″ S and 46° 45′ 00″ E) and in Matanga, district region of Vangaindrano, located on the east coast of Madagascar (23° 31′ 00″ S and 47° 33′ 00″ E) (Fig. 1). Ankazomborona is in the tropical stratum area with a seasonal, endemic and low transmission defined as 10-50 cases per 1000 population per year. While, Matanga is in equatorial stratum area with a perennial endemic and high transmission defined as more than 100 cases per 1000 population per year [20]. Malaria prevalence by microscopy among children 6-59 months of age was 9.0 and 8.8% respectively, in the zones encompassing Ankazomborona and Matanga in 2016 [21]. Plasmodium-positive RDT (SD BIOLINE Malaria Ag P.f/Pan, 05FK63) were collected. They were obtained from children aged 6 months to 15 years suffering to uncomplicated malaria. These RDT containing blood samples were conserved at room temperature before DNA extraction.

DNA extraction
To improve the protocol aiming at extracting parasite DNA from RDT, EDTA tubes containing whole blood infected by P. falciparum and P. vivax were used. Blood smears were read to identify Plasmodium species and estimate the parasite density (parasites/µL). Blood samples were then diluted with non-infected blood to obtain aliquots containing parasitaemia ranging from 1500 to 5 parasites/µL. Four RDT were tested for each dilution. The cassettes of RDT were opened laterally and the strips were taken out and cut for DNA extraction. For each parasite densities, four different parts of the strip were used to estimate the best yield of DNA extract: (A) distal transmission levels and provide new data for guiding the implementation of local strategies for malaria control and elimination. Two methods of DNA extraction were applied: the Instagene Matrix © method (BioRad ™ ), according to the manufacturer's instructions and a simple elution method in water as previously described [4].

Plasmodium falciparum and Plasmodium vivax detection by nested-PCR
Nested-PCR [22] was performed to detect P. falciparum and P. vivax DNA and estimate which DNA extraction method and which part of RDT strip provide the more reliable results. Nested-PCR assays showed that all parts of RDT strip and simple elution method in water were the best approaches (see "Results" section). These methods were then selected for all further experiments.

Plasmodium falciparum msp1, msp2 and glurp genotyping
The polymorphic region of msp1, msp2 and glurp were genotyped using nested-PCR. Primers targeting the block 2 region of msp1, the block 3 region of msp2, and the RII repeated region of glurp were used for primary PCR ( Table 1). All PCR reactions were carried out in a total volume of 25 μL, containing 200 nM dNTP mix, 2 mM MgCl 2 , 200 nM each of forward and reverse primers for both msp1, msp2 and glurp, 0.5 U of Taq DNA Polymerase (Bioline) and 3 μL of extracted DNA, used as template. PCR amplification of msp1, msp2 and glurp genes comprised an initial step of 94 °C for 5 min followed by 30 cycles of 94 °C for 30 s, 55 °C for 30 s, 72 °C for 1 min 30 s, and a final extension of 72 °C for 5 min.
For the nested-PCR, specific primers to allelic families of msp1 (MAD20, K1, and RO33), and msp2 (3D7 and  (Table 1). This secondary reaction contained the same reagents as the primary reaction except primers (Table 1) and 2 μL of the primary PCR product was used as DNA template. The cycling profile for the secondary PCR was similar to the primary PCR for glurp while for msp1 and msp2 the annealing temperature was increased from 55 to 60 °C. DNA from non-infected blood and from reference P. falciparum strains (3D7, Dd2 and 7G8) were included in each set of PCR reactions as a negative and positive controls.
Eight microlitres of nested PCR products were loaded on 2% agarose gel stained with ethidium bromide and separated by electrophoresis for an average of 60 min at 120 V. After electrophoresis, the gels were visualized under UV trans-illumination using Image Lab gel doc system and then analysed to estimate the bands sizes. PCR products size were estimated by Image Lab software using 100 bp DNA ladder marker.
Polyclonal infection was defined by the presence of more than one allele for a given gene [23,24]. Multiplicity of infection (MOI) was defined by the number of genotypes per infection [25].

Optimization of DNA extraction
Out of 16 RDT samples (8 tested with P. falciparum infected bloods and 8 by P. vivax infected bloods), DNA extracted from all parts of RDT strip and simple elution method in water gave successful results by PCR amplification (Fig. 3).

Genetic diversity, allelic frequency and multiplicity of infection
Generated alleles were classified according to their fragment sizes for msp1, msp2 and glurp. Allelic frequency for an allele was defined as being its proportion compared to the total number of all detected alleles in the whole isolates. Plasmodium falciparum isolates in both sites were highly polymorphic, since almost alleles detected were once. For msp1 gene, 14 different alleles were observed: 4 alleles for MAD20 (size range 150-250 bp), 1 allele for RO33 (size range 120-150 bp) and for 6 alleles for K1 (size range 140-300 bp) ( Table 3). Only one sample from Matanga presented polyclonal infection for MAD20 allelic family (MAD20-D) ( Table 3). For msp2 gene, only the FC27 allelic family was observed but the genetic diversity was high: 6 different alleles were detected among the 9 positive samples (Table 4). For glurp gene, out of the 38 P. falciparum positive samples, 20 (52.6%; 95% CI 36.0-68.7%) were successfully amplified. As for msp1 and msp2 genes, a high proportion of different alleles (N = 8) was observed (Fig. 4A, B, Table 5).

Used RDT as source of DNA for molecular analysis
This preliminary study, focused on Plasmodium species detection and msp1, msp2 and glurp genotyping, was performed to evaluate the feasibility of using used RDT, collected from public health facilities in Madagascar, as source of DNA for molecular analysis. Previous findings revealed that used RDTs provide a valuable source of parasite DNA, which can be used for molecular analysis [4-6, 26, 27]. Two DNA extraction methods, Instagene Matrix © method (BioRad ™ ) and simple water elution method, were tested in the present study. Compared methods for DNA extraction showed that simple water elution method gave higher successful result. Thus, this method was adopted in the study. One hundred seventy  used RDT samples were analysed. The proportion of Plasmodium infection was 23.5% (95% CI 17.5-30.8%) including 92.2% P. falciparum positive, 5% P. vivax positive and 2.5% P. falciparum/P. vivax mixed positive samples. Although this low proportion, this study highlighted the opportunity using used RDT for malaria diagnosis and molecular analysis [4]. Furthermore, studies confirmed that they also can be useful for molecular markers surveillance and whole genome sequencing [28,29]. The main advantage is that no other consumable, nor reagent for blood sampling are needed and RDT are widely available in public health facilities since 2008 in Madagascar [30]. Nevertheless, the lack of DNA quality and quantity analysis constitute one of the limitations for the present study. Furthermore, the parasitaemia for each samples are not available and the laboratory analysis were carried out 8 months after the collection of used RDT that were stored two months in field. Thus, the storage of used RDT in the field conditions might have likely affected the DNA integrity. DNA extraction method may should be improved for increasing the yield of parasite DNA [5]. Yet, several parameters such as parasitaemia and the impact of storage duration and conditions of used RDT on the quality of the DNA extract should be considered. Besides, in order to increase the power of the study, conducting analysis with large samples is needed.
Due to the low amplification rate and the limited samples, only the FC27 allelic family was detected for the msp2, concordant with data observed by Rakotomanjaka in the same locality (Matanga) in 2016 [41]. This is not the case in all localities in Madagascar, since several  studies using DNA sequencing method, reported higher frequency of 3D7 allelic (up to 48%) such as in Andapa, Mahasolo and Saharevo [42,43] [44]. Whereas, the study was carried out during the peak of malaria transmission [45]. Analysis based on the PCR product fragments showed 4 distinct alleles for MAD 20 allelic family (11 samples), 1 allele for RO33 (10 samples) and 6 alleles for K1 (6 samples). The K1 allelic family was found at low frequency (28.6%) compared to MAD20 and RO33, but this allelic family was more polymorphic (6 distinct alleles for 6 samples). For FC27 allelic family of msp2, out of the 9 positive samples, 6 distinct alleles were detected. Regarding to glurp gene, 8 distinct alleles were observed in the 20 positive samples. Polymorphism analysis showed high genetic diversity of P. falciparum populations in Ankazomborona and Matanga. Majority of alleles detected during in this study were single with a high polymorphism level. No significant difference of polymorphism level was detected in two study sites. Parasite genetic diversity is one of key elements to consider for strategies implementation to fight against malaria [40]. It has an impact on parasite transmission and control strategies, and it varies according to transmission in endemic areas [7][8][9][10]. According to the results, it seems, transmission intensity is similar in Ankazomborona and in Matanga. However, parasites transmission and genetic diversity could be higher during the peak of malaria transmission and may reflect the dynamics of parasites transmission [23,25]. Hence, a study on large samples is crucial before giving final conclusion.
This study showed 28.6% multiple infection for P. falciparum msp1 gene not for msp2 and glurp. These results indicate that polyclonal infection frequency is low compared to those found in other countries [46,47]. The MOI average value was 1.79 ± 0.74, a lower value than in other northwestern Ethiopia (2.6), in Republic of Congo (2.64), and in Equatorial Guinea (5.5) [25,37,40]. MOI for Ankazomborona and Matanga did not differ significantly. Thus, hypothesis concerning the link between the high level of MOI and malaria severity, particularly in areas with high transmission rate [11], is not verified during this study due to the limited number of samples. Despite this, the present study revealed that P. falciparum populations circulating in Madagascar showed a high level of genetic polymorphism. That high polymorphism reflects transmission intensity. Consequently, it would be interesting to increase samples number in order to continue genetic diversity study over time and space for carrying out a cartographic study about Plasmodium allelic families' distribution in Madagascar.

Conclusion
The present study highlighted the possibility of performing malaria diagnosis and molecular genetic diversity analysis of Plasmodium using blood samples collected on RDT as DNA source. Preliminary data from this study showed high genetic diversity of P. falciparum populations in Ankazomborona and in Matanga but multiple infection rate was low in the both study sites. This study provides fundamental information on P. falciparum genetic diversity and allow to update available data. The results of this study can be used as a baseline information for future studies on parasite transmission dynamics and to evaluate effectiveness malaria prevention and control strategies.