Assessment of drug resistance associated genetic diversity in Mauritanian isolates of Plasmodium vivax reveals limited polymorphism

Background Plasmodium vivax is the predominant malaria species in northern Mauritania. Molecular data on P. vivax isolates circulating in West Africa are scarce. The present study analysed molecular markers associated with resistance to antifolates (Pvdhfr and Pvdhps), chloroquine (Pvmdr1), and artemisinin (Pvk12) in P. vivax isolates collected in two cities located in the Saharan zone of Mauritania. Methods Blood samples were obtained from P. vivax-infected patients recruited for chloroquine therapeutic efficacy study in 2013 and febrile patients spontaneously consulting health facilities in Nouakchott and Atar in 2015–2016. Fragments of Pvdhfr (codons 13, 33, 57, 58, 61, 117, and 174), Pvdhps (codons 382, 383, 512, 553, and 585), Pvmdr1 (codons 976 and 1076) and Pvk12 (codon 552) genes were amplified by PCR and sequenced. Results Most of the isolates in Nouakchott (126/154, 81.8%) and Atar (44/45, 97.8%) carried the wild-type Pvdhfr allelic variant (IPFSTSI). In Nouakchott, all mutants (28/154; 18.2%) had double Pvdhfr mutations in positions 58 and 61 (allelic variant IPFRMSI), whereas in Atar only 1 isolate was mutant (S117N, allelic variant IPFSTNI). The wild-type Pvdhps allelic variant (SAKAV) was found in all tested isolates (Nouakchott, n = 93; Atar, n = 37). Few isolates in Nouakchott (5/115, 4.3%) and Atar (3/79, 3.8%) had the mutant Pvmdr1 allele 976F or 1076L, but not both, including in pre-treatment isolates obtained from patients treated successfully with chloroquine. All isolates (59 in Nouakchott and 48 in Atar) carried the wild-type V552 allele in Pvk12. Conclusions Polymorphisms in Pvdhfr, Pvdhps, Pvmdr1, and Pvk12 were limited in P. vivax isolates collected recently in Nouakchott and Atar. Compared to the isolates collected in Nouakchott in 2007–2009, there was no evidence for selection of mutants. The presence of one, but not both, of the two potential markers of chloroquine resistance in Pvmdr1 in pre-treatment isolates did not influence the clinical outcome, putting into question the role of Pvmdr1 mutant alleles 976F and 1076L in treatment failure. Molecular surveillance is an important component of P. vivax malaria control programme in the Saharan zone of Mauritania to predict possible emergence of drug-resistant parasites.

Background Both Plasmodium falciparum and Plasmodium vivax are endemic in Mauritania with approximately 300,000 malaria cases reported in 2017 [1]. In the northern Saharan zone of Mauritania, particularly in Nouakchott, the capital city of the country, and Atar, the provincial capital of Adrar region, malaria transmission is seasonal, and P. vivax has been shown to be the main causative species of malaria parasite [2][3][4]. Infections caused by P. vivax have been considered for a long time as benign. Recently, several prospective studies recognized that P. vivax may be responsible for significant morbidity and even severe disease leading to mortality in endemic areas [5]. In the absence of an effective malaria vaccine, the use of insecticide-impregnated bed nets and antimalarial drug administration remain the only ways to prevent or treat the disease and reduce the probability of transmitting the parasite to Anopheles mosquitoes [6]. However, in many malaria-endemic countries, including Africa, the control strategy based on anti-malarial drugs has been facing the problem of parasite strains resistant to chloroquine, amodiaquine, antifolates, and mefloquine. In some countries, artemisinin resistance, including resistance to artemisinin-based combination therapy (ACT) dihydroartemisinin-piperaquine, has also emerged and spread [7][8][9][10]. To prevent, delay, and overcome the emergence of multidrug-resistant parasites, ACT has been recommended since 2001 for the treatment of uncomplicated malaria in endemic countries [11].
The first molecular surveillance of P. vivax in Mauritania was conducted with isolates collected in [2007][2008][2009] in Nouakchott, the capital city [3]. The present study is a follow-up molecular surveillance of P. vivax resistance in Nouakchott and Atar, both of which are situated in the Saharan zone of Mauritania, using P. vivax isolates collected between 2013 and 2016. As in the earlier study [3], polymorphisms in Pvdhfr, Pvdhps, and Pvmdr1 were analysed in the present study. In addition, P. vivax kelch propeller domain located on chromosome 12 (Pvk12), which was shown to be homologous to Pfk13 in 2015, was analysed in the present study [39]. Pvcrt-o gene was not sequenced in this study because there is no association between Pvcrt-o mutations and chloroquine resistance [31,32].

Patients and blood sample collection
Blood samples used in the present study were collected from P. vivax-infected patients recruited for chloroquine therapeutic efficacy study conducted in Nouakchott and Atar in 2013 [40] and febrile patients spontaneously consulting health centres and hospitals and screened for malaria using rapid diagnostic test for malaria (SD Bioline P. falciparum histidine-rich protein 2 and P. vivax plasmodial lactate dehydrogenase antigen rapid diagnostic test; Standard Diagnostics, Inc., Yongin, Republic of Korea) in Nouakchott and Atar in 2015-2016 ( Fig. 1). Two or three drops of fingerprick capillary blood (approximately 150-200 µL) were spotted directly on Whatman 3MM blotting paper (GE Healthcare Life Sciences, Bucks, UK), dried, and stored in individual sealed plastic bag at − 20 °C until use.

PCR and sequencing of drug resistance genes
Parasite DNA was extracted from Whatman ® 3MM papers by Chelex ® -saponin method [41]. Plasmodium species was identified by nested PCR targeting cytochrome b gene [42]. The presence of Pvk12 mutation (V552I) observed in Cambodian P. vivax isolates was analysed using PCR protocol described by Popovici et al. [39].
Mutations associated with drug resistance in Pvdhfr, Pvdhps, and Pvmdr1 genes were investigated using slightly modified nested PCR protocols described earlier [3]. The sequences of specific primers are presented in Table 1. The reaction mixture for the primary PCR amplification consisted of 10 µL of DNA template, 0. Nested PCR amplifications were performed using a thermal cycler (Eppendorf 950000015 Mastercycler Gradient Thermal Cycler; Hinz GmbH, Hamburg, Germany). The following program was used to amplify Pvdhfr, Pvdhps, and Pvmdr1 genes: initial denaturation at 95 °C for 5 min followed by 45 cycles of denaturation at 95 °C for 1 min, annealing at 53 °C for 1 min and elongation at 72 °C for 1 min, and the final elongation step of 72 °C for 10 min. For the secondary amplification, 2 μL of the primary PCR product was used as DNA template. The thermal cycling program and PCR mixture were identical to those of the primary PCR, except for the annealing temperature which was set at 65 °C and the volume of DNA template.
The products of the secondary PCR were visualized by electrophoresis on a 1% agarose gel. The PCR products were purified using Wizard ® SV gel and PCR clean-up system (Promega Corp, Fitchburg, WI) according to the manufacturer's instructions. DNA sequencing of the PCR products was performed by Eurofins Genomics (Les Ulis, France) using BigDye Terminator v3.1 cycle sequencing kit (Applied Biosystems Courtaboeuf, France) and ABI 3730XL sequencer. Sequences were read using Chromas DNA sequencing software (Technelysium Pty Ltd., South Brisbane, Australia) and aligned using Clustal X2 software (Conway Institute, University College Dublin, Dublin, Ireland) to search for SNPs. Polymorphisms were identified in both forward and reverse strands by comparing the sequences to the reference sequences of Pvdhfr (GenBank accession no X98123), Pvdhps (GenBank accession no AY186730), Pvmdr1 (Genbank accession no AY618622), and Pvk12 (PVX_083080).

Results
Among blood samples collected in Nouakchott and confirmed positive by rapid diagnostic test, the molecular markers of drug resistance were analysed for the following numbers of isolates: 154 for Pvdhfr, 93 for Pvdhps, 115 for Pvmdr1, and 59 for Pvk12 (

Discussion
In Mauritania, the treatment of uncomplicated malaria had been based on chloroquine and sulfadoxinepyrimethamine (SP) as the first-and second-line drugs until 2006, respectively. Following the emergence and spread of chloroquine-resistant P. falciparum in West Africa [43], the Mauritanian health authorities adopted, in 2006, a new therapeutic strategy for the management of malaria cases based on ACT (using either artesunateamodiaquine or artemether-lumefantrine) as the firstline treatment for all malaria cases without distinction of Plasmodium species. This national drug policy was adopted to conform to the WHO guidelines for African countries and was not a decision based on previous clinical studies demonstrating chloroquine-resistant P. falciparum in Mauritania.
Unlike P. falciparum, which is the predominant Plasmodium species in Africa, molecular data on P. vivax are scarce to non-existent in most West African countries. The Saharan zone of Mauritania is an exception [44,45]. The results of the present study showed that mutations in Pvdhfr, Pvdhps, Pvmdr1 and Pvk12 occur in some P. vivax isolates collected recently in the Saharan zone of Mauritania. It was also observed that in Pvdhfr, a marker known to accumulate mutations at specific codons and increase the level of resistance to pyrimethamine, the number of mutations in a mutant isolate was limited to two, and the key Pvdhfr S117N substitution was observed in a single isolate from Atar. These results are in general agreement with the previous study conducted in isolates collected in 2007-2009 in Nouakchott [3]. In that study, only 12% (10 of 86) of isolates   were mutants carrying double mutations S58R and S117N, and, as in the present study, none of the isolates analysed earlier (n = 94) had mutations in Pvdhps. For more than a decade, the use of SP has been restricted to pregnant women for intermittent preventive treatment (IPTp) in Mauritania, as well as in children for seasonal malaria chemoprevention (amodiaquine + SP) in some African countries in the Sahelian zone. Both of these preventive strategies are directed against P. falciparum. The use of SP to treat P. vivax malaria is not recommended due to its inherent resistance or lower susceptibility to antifolates [46]. In practice, some pregnant Mauritanian women under IPTp in the Saharan zone are exposed to the risk of P. vivax infection. Although prospective clinical studies would be required to confirm the benefit of IPTp even against P. vivax, molecular data tend to support SP efficacy against P. vivax.
The earlier molecular study in Nouakchott showed that most isolates (75/103, 73%) had Y976 wild-type Pvmdr1 allele, as in the present study (110/115, 95.7%) [3]. However, that study showed a high proportion (98%) of mutant F1076L allele in Pvmdr1, whereas in the present study, mutation was absent in codon 1076. The origin of this discrepancy is unknown. This observation calls for further molecular monitoring in Nouakchott. Pvmdr1 Y976F and F1076L mutations were suggested to be associated with chloroquine and amodiaquine resistance [33,37,38]. However, several in vitro and clinical studies have failed to confirm the association between Pvmdr1 mutations and drug resistance [47][48][49][50][51][52][53]. In the present study, among 115 and 79 isolates with Pvmdr1 sequences in Nouakchott and Atar, 51 and 48 were pre-treatment blood samples from patients treated with chloroquine in 2013, respectively. In many patients enrolled in chloroquine therapeutic efficacy studies, parasitaemia was cleared on or before day 3, and none of the patients had parasitaemia on day 7 [40]. The outcome was adequate clinical and parasitological response in all patients on day 28. These results suggest that the presence of one of the two mutant Pvmdr1 alleles (i.e., 976F or 1076L, but not both) is not associated with chloroquine treatment failure in Nouakchott and Atar. It remains unclear whether both 976F and 1076L are required for chloroquine treatment failure since none of the isolates analysed in the present study carried the double mutations in Pvmdr1. Further clinical and molecular studies are required to assess the relevance of Pvmdr1 mutations in chloroquine resistance, in particular in African countries, such as Ethiopia, where chloroquine-resistant P. vivax occurs [54][55][56][57].
Artemisinin-resistant P. falciparum occurs in Southeast Asia, and resistance seems to be conferred by specific mutations in the propeller domain of Kelch 13 (K13; PF3D7_1343700) [27]. In P. vivax, Pvk12 was identified as the homologous gene of Pfk13 [39]. At present, there is no molecular indication that Pvk12 undergoes a high degree of mutations as its P. falciparum homolog, even in Southeast Asia [39,[58][59][60]. The sequence data of Mauritanian P. vivax isolates also indicate the absence of polymorphism in the Pvk12 gene fragment that was amplified and analysed. Moreover, clinical and in vitro studies carried out so far do not suggest artemisinin resistance in P. vivax. Further molecular surveillance is warranted to anticipate the possible emergence of artemisinin-resistant P. vivax around the world.

Conclusions
Few Mauritanian P. vivax isolates were characterized to carry mutations in Pvdhfr, Pvdhps, Pvmdr1, and Pvk12 markers. The molecular data are in general agreement with the high clinical efficacy of chloroquine previously demonstrated in Nouakchott and Atar and with an earlier study on isolates collected in 2007-2009 in Nouakchott, suggesting that mutations in these molecular markers of drug resistance are not being selected. Due to the unavailability of chloroquine in the official drug outlets of the country, P. vivax infections have been treated with ACT over more than a decade. This may explain, at least in part, the absence of selection of mutant P. vivax parasites. Molecular surveillance of P. vivax is an important component of malaria control in northern Mauritania where this parasite species is predominant.