Analyzing Deoxyribose Nucleic Acid from Malaria Rapid Diagnostic Tests to Study Plasmodium falciparum Genetic Diversity in Mali

Abstract

We evaluated the use of positive malaria rapid diagnostic tests (mRDTs) to determine genetic diversity of Plasmodium falciparum in Mali. Genetic diversity was assessed via multiple loci variable number of tandem repeats analysis (MLVA). We performed DNA extraction from 104 positive and 30 negative used mRDTs that had been stored at ambient temperature for up to 14 months. Extracted DNA was analyzed via quantitative polymerase chain reaction (qPCR), and MLVA genotyping was then assessed on positive qPCR samples. Eighty-three of the positive mRDTs (83/104, 79.8%) and none of the negative mRDTs were confirmed P. falciparum positive via qPCR. We achieved complete genotyping of 90.4% (75/83) of the qPCR-positive samples. Genotyping revealed high genetic diversity among P. falciparum populations in Mali and an absence of population clustering. We show that mRDTs are useful to monitor P. falciparum genetic diversity and thereby can provide essential data to guide malaria control programs.

Genetic diversity analysis of Plasmodium falciparum in humans and parasite population dynamics are useful to monitor malaria control and elimination strategies. Indeed, several multiple loci variable number of tandem repeats analysis (MLVA) studies using microsatellite markers have shown that P. falciparum population genetics correlated with malaria transmission intensity in Africa and provided epidemiologically relevant information.^1–5 Assessing malaria parasite genetic diversity over time can also help to evaluate malaria control programs. For instance, a P. falciparum microsatellite-based survey in Djibouti has revealed a significant decline in genetic diversity between 1998 and 2009, which was compatible with pre-elimination goals.⁵

Plasmodium falciparum genetic diversity has been well described in humans throughout the world^1–5; however, this type of study is difficult to manage in the field. Indeed, collection, transportation, and storage of large amounts of blood samples remain difficult in remote areas with a tropical climate. Sampling of dried blood spots on filter papers partially circumvents these technical difficulties, thereby enabling detection and genotyping of P. falciparum from archived specimens.^1,4,5 Nevertheless, patient consent is required as long as specific blood sampling is required for the study.

Malaria rapid diagnostic tests (mRDTs) detect the presence of circulating P. falciparum-specific antigens, such as histidine-rich protein 2 (PfHRP2) and lactate dehydrogenase. They are highly recommended by the World Health Organization for systematic malaria diagnosis before artemisinin-based combination therapy of uncomplicated malaria cases.⁶ As mRDTs are already used for case management, they are widely available in malaria-endemic countries and therefore represent a potential source of P. falciparum DNA for large population studies. Furthermore, patient consent is not required for such investigations, as mRDTs are passively collected during patient care and would normally been discarded once interpreted. Previous studies have shown that used mRDT nitrocellulose strips harbor P. falciparum DNA,^7–10 and thereby enable the detection of antimalarial drug-resistant genes via the single nucleotide polymorphism genotyping.⁸ This study aimed to assess whether used mRDTs stored at room temperature provide sufficient quality DNA to conduct MLVA genotyping of P. falciparum in Mali.

Between October 2013 and January 2015, we randomly selected a total of 134 used mRDTs (104 PfHRP2 positive and 30 PfHRP2 negative) from four sites in Mali (Figure 1 , Table 1). Blood samples were collected via finger prick (5–10 μL) and absorbed onto mRDTs (SD BIOLINE Malaria Ag P.f^® and P.f/Pan^®, Standard Diagnostics, Kyonggi, Republic of Korea) by Malaria Research and Training Center (MRTC) clinicians during the systematic testing of symptomatic febrile patients. Used mRDT samples were stored and transported at ambient temperature to the MRTC Central Laboratory in Bamako, Mali. The samples were then randomly selected and air transported to Marseilles, France, where P. falciparum DNA was extracted from January 2015 to August 2015.

Figure 1.

Map of the four study sites in Mali. The size of the red dots indicates the relative approximate number of inhabitants of Bamako, Bougoula, Djoliba, Samako, and Kolle. Samako and Kolle were pooled as a single site because of geographic proximity.

Table 1.

Epidemiological and sampling characteristics as well as the qPCR and MLVA results of the 104 mRDT-positive samples collected from four sites in Mali

Study sites	Djoliba	Kolle (K) Samako (S)	Bamako	Bougoula	Total
Sampling period	October 2013 to January 2014	November 2014 (K)	December 2014	January 2015
		December 2014 (K)	January 2015
		January 2015 (K) (S)
Endemicity*†	Hyperendemic	Hyperendemic	Hypoendemic	Hyperendemic
Patients sampled, no.	34	27	28	15	104
Positive PCR for Plasmodium falciparum, n (%)	29 (85.3)	24 (88.9)	20 (71.4)	10 (66.7)	83 (79.8)
Complete genotype (N = 8 loci), n (%)	28 (96.5)	21 (87.5)	18 (90.0)	8 (80.0)	75 (90.4)
Genetic diversity (mean ± SD)	0.76 ± 0.18	0.75 ± 0.16	0.78 ± 0.17	0.69 ± 0.36	0.75 ± 0.22

MLVA = multiple loci variable number of tandem repeats analysis; mRDTs = malaria Rapid Diagnostic Tests; qPCR = quantitative polymerase chain reaction; SD = standard deviation.

^*Data from various Malian official reports and studies reviewed in a published thesis in French. (Doumbo O, 1992. Epidémiologie du paludisme au Mali, étude de la chloroquinorésistance, essai de stratégie de contrôle basée sur l'utilisation de rideaux imprégnés de permethrine associée au traitement systématique des accès fébriles. Thèse de Doctorat. Sciences Biologiques, Montpellier II, France.)

^†Malaria endemicity levels based on P. falciparum prevalence among children aged 2–10 years (PfPR_2–10), according to the World Health Organization classification: hypoendemic, 0–10%; hyperendemic, 50–75%.

The mRDT nitrocellulose strips (cut into five identical pieces) were incubated for 48 hours at ambient temperature in 800 μL of lysis buffer (bioMérieux, Marcy l'Etoile, France). The extracted DNA was eluted in 100 μL of elution buffer using a NucliSENS EasyMAG instrument (bioMérieux).¹¹ Screening for P. falciparum was performed via quantitative polymerase chain reaction (qPCR) using a LightCycler 480 PCR system (Roche Diagnostics, Meylan, France) with specific primers targeting the 18S rRNA gene.¹² Each experimental run included both a negative (no template) and a positive (P. falciparum 18S rRNA plasmid) control. Standard curves were generated with serial 10-fold dilutions of the plasmid to allow for species-specific quantification of parasite density (number of parasites/μL of blood). We assumed that each genome of P. falciparum has five copies of the 18S rRNA gene (as observed in the 3D7 genome).¹³ The qPCR-positive samples were genotyped using eight polymorphic microsatellite markers specific for P. falciparum (Poly α, TA109, TA1, TA81, TA42, ARA2, PfPK2, and Pfg377).¹ Microsatellite amplification was performed applying a semi-nested PCR strategy using fluorescent end-labeled primers.¹ PCR products were analyzed using an ABI 3130XL capillary sequencer (Applied Biosystems, Foster City, CA). To differentiate allele peaks from stutter peak artifacts, we scored multiple alleles per locus if minor peaks were > 33% of the height of the major peak, corresponding to the predominant allele.¹ We also discarded peaks with a fluorescence intensity < 100 units. Genetic diversity metrics were assessed using Arlequin v3.5 (Excoffier & Lischer 2010) software based on complete genotypes. For a haploid organism, genetic diversity was defined as a measure of the probability to randomly draw a pair of different alleles from an allelic pool. Potential values ranged from 0 (no diversity, 100% similarity between alleles) to one (maximal diversity, 100% of the alleles are different). Because of the frequent occurrence of multiclonal infections, we only considered major peaks when determining genotypes for genetic diversity and haplotype analysis.^1–5 To analyze the relationship between P. falciparum haplotypes, we computed a dendrogram using the Unweighted Pair Group Method with Arithmetic Mean (UPGMA) method (BioNumerics v7.1 Software, Applied Maths, Ghent, Belgium).

Eighty-three of the positive mRDTs (83/104, 79.8%) and none of the negative mRDTs were confirmed P. falciparum-positive via qPCR (Table 1). Among the 83 qPCR-positive samples, 75 (90.4%) were successfully genotyped for all test loci (Table 1). The eight remaining samples displayed one (N = 3), two (N = 2), three (N = 1), and four (N = 2) negative loci of the eight loci tested (see Table 3 for complete microsatellite data). Mean PCR cycle threshold (C_t) values were 31.94 ± 3.15 for the 75 samples that were completely genotyped (estimated parasitemia ranging from 11 to 512,000 parasites/μL) versus 37.52 ± 1.92 for the eight incompletely genotyped samples (estimated parasitemia ranging from 0.01 to 3,620 parasites/μL) (Table 2).

Table 3.

Microsatellite data of the 83 genotyped samples

Site	Sample	Poly α				TA109				PfPK2			ARA2				Pfg377			TA42			TA81			TA1
Kolle	T10	166	163	157		183				168	171		76	82			104	107		192			125	128		168	180
Kolle	T11	157				171				180			67				104			207			143			180
Kolle	T12	160				168				165			76				107			207			125			174
Kolle	T13	157	136			180				180			64				104			192			137			180
Kolle	T15	157				192				168			73				104			192			131			180	168
Kolle	T16	157				183				165			70				104			192			131			171
Kolle	T17	154				183	195			180			67				104			192			131	137		168	189
Kolle	T18	169				183	171			165	183		64	70			104			183			131			183	180
Kolle	T19	148	166			183				171	168	174	82	70	67		107			207	192		125	131		189	174	183
Kolle	T129	163				180				177			67				107			192			134			171
Kolle	T130	163				180				168							107			192			125			165
Kolle	T131	163				213				177			82				107			192			131			174
Samako	T20					168				180	171						107			192			128
Samako	T21	163				168	183			174	168		70	67			107			192			131	125		180
Samako	T22	163	157			180				174			82				107			207			128			174
Samako	T23	169	172	157		168	180			171			70	64			104	107		192			134	143		168	177	189
Samako	T24					180	198			174							104			192			128			189	186
Samako	T25	151	160			183				174	171		67	79			104			192			128	131		165	168
Samako	T26	157				168				174	171		64	79			107			192			128	125		183	180
Samako	T28	154	157			186	201			177	174		73				104			192			137			168	171
Samako	T124	157	142	166	175	168	183	180	198	168	171		70	64	67	82	107			192	207		125	131		165	171	189	198
Samako	T125	109				168	182			168	174	180	82	73			104	113		192			131	137		171	189
Samako	T126	157				168				177			70				107			192			131			180
Samako	T127	157	172			189	180	168		165	195		73	64			107			207	192	183	125			183	165
Djoliba	T29	178				180				177			73				107			192			131			177
Djoliba	T30	169				168				174			79				104			207			131			165
Djoliba	T31	154				186				189			73				107			192			128			174
Djoliba	T32	169	154			171				186	168		76				107			192			137	125		180
Djoliba	T33	160	157			168	180			171	165		76				104	107		192			134	128		177	165
Djoliba	T34	163				183				192			94				107			207			131			174
Djoliba	T35	172	184			168	174			171	195		70				107			192	189		128			174
Djoliba	T36	157				180	183			171	165		70				104			207	222		131	137		171	183
Djoliba	T37	163				204				171			85				107			207			122			174
Djoliba	T38	163	184			195	183	171		174	177		73				107			192			128			171
Djoliba	T39	115	160			159				168	171		70				104			192			125	170		174
Djoliba	T40	154	166			153				171			76	73			113			207			131			171
Djoliba	T41	178				183				168			73				107			192			125	128		174
Djoliba	T42	154				207	171			168			94	64	73		107			192			131			183	171
Djoliba	T43	169	160			180				180	165		79				107			207			125			171
Djoliba	T44	175				171				165			79				104			192			125			183
Djoliba	T45	184				189				174			70				101			192			125			174
Djoliba	T46	157				183							73							207			140			174
Djoliba	T47	148				180				171			73				110			192			125			168
Djoliba	T48	160				171				165	171		79	70			101	107		195			143	134		174
Djoliba	T50	157	160	163		183	171	186	204	168	171	174	76	67	70		104	101	107	192			131	128	134	165	174	168	177
Djoliba	T51	166				171				174			70				107			192			140			165
Djoliba	T52	157				183	168			183	177	171	76	67			107			192			125			168
Djoliba	T54	154				180				198	162		76				104			192	207		128			171
Djoliba	T55	160	154			168				168	177		76				107	104		192			125	128	134	171	168
Djoliba	T56	169				162				186	177		70				107			192	207		128			189	177
Djoliba	T57	157				168				171			73				107			192			116	128		165
Djoliba	T58	160	178			177	189			174			70	64			107			192	207		125			168	174
Djoliba	T59	142				183	189			168			76	67			107			192			125			177	183
Bougoula	T133	160				183				186			76				107			192			116			171
Bougoula	T134	157				201																	128	125		132
Bougoula	T135	151	154			180	183			192			85	73			107			192			125			171
Bougoula	T137	181				171				171			82				107	113		192	204	207	128			168
Bougoula	T138	166	154			168	183			168	180		76	70	82		107	92		192			125			171
Bougoula	T139	157	166			177	168			168			76	70			107			192			128			165	174
Bougoula	T140	151				171	168	180		171			70				107			192	207		134	131		177	180	174
Bougoula	T141	160	157			165	192			168	165		82				107			192			131	134	128	174
Bougoula	T144	151				189				174			70				104			192			122			183
Bougoula	T146					168				186			64				107			192			128	134		171
Bamako	T1	169				177				174			82				104			192			134	128		171
Bamako	T3	154	178			189				174			70				104			207			128			201
Bamako	T6	166				165				168			70	82			104			192			131			186
Bamako	T7	139				180				165			70				110			207			128			168
Bamako	T9	154				189				168			67				98			192			134	128		174
Bamako	T155					180				171			76				107			192			140			171
Bamako	T156	157				171				168			70	73			107			192			134			186
Bamako	T157	157				168				171			79				104			192			131			165
Bamako	T159	169				168				174			76				107			207			131			180
Bamako	T160	139				171				165			70	73			110			207			125	131		174	165
Bamako	T162	172				183	186			171	162		64	73			101	107		192			125	134		177
Bamako	T163	172				183				171			76				104			192			128			177
Bamako	T171	178				168				165			70				110	95		192			128			174
Bamako	T172	157				171	180			174			67	73			104	110		192			131			168	180
Bamako	T174	166				183				171			70				107			192			128			186
Bamako	T176	163				183				171			76				104	89		192			134			192
Bamako	T177	154				183				177													131
Bamako	T178	154	157			168				177			73				107	92		192			125			180
Bamako	T179	160				180				168			79				107	92		192			128			177
Bamako	T180	157	169	148		180	168	183		177	171	180	73	76	85		107			192			125			171	174

The allele length (in base pairs) for each Plasmodium falciparum tested loci (Poly α, TA109, TA1, TA81, TA42, ARA2, PfPK2, and Pfg377) is provided. The empty cells of the table correspond to negative loci.

Table 2.

Quantification of samples via qPCR cycle threshold and estimated parasitemia based on the number of positive loci

	Complete genotypes (8 positive loci; N = 75)	Incomplete genotypes (< 8 positive loci; N = 8)
PCR cycle threshold (mean ± SD)	31.94 ± 3.15	37.52 ± 1.92
Estimated parasitemia* (parasites/μL blood) (mean ± SD)	43,067 ± 82,558	749 ± 1,280
Range of estimated values of parasitemia* (parasites/μL blood)	11–512,000	0.01–3,620

qPCR = quantitative polymerase chain reaction; SD = standard deviation.

^*Estimation of parasite density: ([copy number of the gene in 1 μL of DNA] × [100/25]). We assumed that each genome of Plasmodium falciparum has five copies of the 18S rRNA gene,¹³ that 5 μL of blood was spotted onto malaria rapid diagnostic tests, and that extracted DNA was eluted into 100 μL.

Genetic diversity was high (mean value of 0.75) and did not differ between the urban site of Bamako and the rural sites of Samako/Kolle (which were pooled together because of geographical proximity) and Djoliba (Table 1). The dendrogram based on the analysis of the 75 complete haplotypes (Figure 2 ) demonstrates that each haplotype was unique, which is consistent with the high level of isolate genetic diversity. The absence of major haplotype clustering between and within study sites indicates a lack of local clonal expansion of P. falciparum.

Figure 2.

UPGMA dendrogram displaying the relationship between the Plasmodium falciparum haplotypes (N = 75). The isolates are colored according to geographic location in the dendrogram. Red: Djoliba; blue: Samako/Kolle; green: Bamako; and gold: Bougoula.

Our findings show that mRDTs are a very accessible, convenient, and efficient tool to detect and genotype P. falciparum from patient blood samples. As mRDTs are already widely used for malaria case management, such samples can be collected and stored in large-scale in malaria-endemic countries. Systematic collection of used mRDTs can be considered for genetic diversity studies without specific population sampling procedures or patient consent requirements. Malaria RDTs are also suitable for field conditions, as they can be transported and stored at ambient temperature, even for extended periods. Indeed, after a 14-month storage period at ambient temperature, DNA quality and quantity were sufficient to perform complete genotyping of 90.4% of the qPCR-positive samples. In addition, mRDTs can be transported in simple packages; this feature is critical as it is increasingly difficult to ship biological samples because of the expansion of international biological risk.

Of 104 mRDT-positive samples, 21 tested negative via qPCR. This discrepancy may be due to false-positive mRDT results associated with the persistence of PfHRP2 antigenemia following malaria treatment, as previously observed.^10,14 Malaria RDT-positive samples that yielded negative qPCR results could also be interpreted as false-negative qPCR results because of the presence of PCR inhibitors or DNA extraction failure. Insufficient parasitic material may also yield a false-negative result, as the blood volume spotted onto mRDTs is low (5–10 μL) compared with the blood volume usually spotted on filter papers (50 μL). However, previous studies have reported the absence of PCR inhibitors in samples obtained from mRDTs.^7,9 Furthermore, the estimated qPCR sensitivity from mRDT samples is much higher than that of mRDT antigen capture and field microscopy (both estimated to 100 parasites/μL).¹⁵ Indeed, the presence of a single parasite in a 5–10 μL blood sample spotted onto an mRDT should be sufficient to yield a positive qPCR result.⁹

Eight of 83 samples (9.6%) were not successfully genotyped for all test loci despite a positive qPCR result. This outcome may be due to low parasite density, as indicated by the higher C_t values (mean C_t of 37) compared with the successfully genotyped samples (mean C_t of 32). This result may also be due to unsuccessful primer recognition because of high polymorphism or local DNA alterations of the target sequence. Nevertheless, 90.4% of samples were completely genotyped for all eight test loci, which provided sufficient data to examine P. falciparum genetic diversity.

Our study demonstrates high genetic diversity among P. falciparum populations in Mali (mean genetic diversity = 0.75), which indicates a high intensity of malaria transmission in the country. These results are compatible with local malaria epidemiology, as Mali has yet to reach the malaria pre-elimination stage. Microsatellite studies performed in other malaria-endemic countries in Africa have also reported high genetic diversity of P. falciparum, ranging from 0.72 to 0.8.^1–4

These techniques of DNA extraction and genotyping from used mRDTs have already been transferred to the MRTC in Mali and will serve to monitor genetic diversity of P. falciparum in the context of malaria control. This approach will also facilitate the monitoring of drug resistance via genotyping of resistance genes. Using mRDTs, future large-scale P. falciparum genetic studies would be relatively cost-effective and easy to carry out, as they circumvent specific blood sampling, storage, and transportation. This technique thus represents a breakthrough in the capacity to bolster field malaria epidemiological studies.