In Mozambique, the first line for uncomplicated malaria treatment is AL and ASAQ the second line and an alternative in case of contraindications[13]. Artemisinin and its derivatives play an important role in killing Plasmodium falciparum by inhibiting the activity of phosphatidylinositol-3-kinase (PfPI3K) [18, 34] and like any other drug, it may have side effects, but they are poorly expressed [18, 35, 36]. Surveillance of validated and/or resistance-associated SNPs of the aforementioned ACTs is a powerful weapon to control the spread of parasite resistance through early detection of mutations [37–39]. Moreover, these mutations are being associated with the reduction in pfkelch13 function, a protein required for parasite-mediated endocytosis of host haemoglobin in the newly invaded intra-erythrocytic ring stages [7, 39]. Thus, the present study was designed to update the current profile of SNPs in P. falciparum pfk13 associated with partial artemisinin resistance in 3 provinces of Mozambique (Niassa, Manica and Maputo), from April to August 2021.
Our results showed that 66,2% of selected samples were confirmed for P. falciparum by real time PCR, and successfully sequenced for pfk13 gene. No validated mutation for artemisinin resistance was observed in our samples. Furthermore, we detected 16 non-synonymous (G449R (GGT»CGT), V454E (GTA»GAA), E455Q (GAA»CAA), D464Y (GAT»CAT), W470R (TGG»AGG), V494L (GTT»CTT), C532W (TGT»TGG), I543S(ATT»AGT) A578S (GCT»TCT), A578P (GCT»CCT), V581A (GTT»GCT), D464H (GAT»CAT), E509D (AGA»AAA), Q654H (CAA»CAC), Q661H (CAA»CAC), L663I (CAA»CAC)) and 5 Synonymous (S459S, C469C, L598L, G548G, G549G) mutations. However, most of these mutations were due to substitution at one or more bases of the codons. Out of these 21 point mutations, three have been reported in Africa, namely C469C [40, 41], A578S [41–43] and Q661H [41, 44]. The remaining point mutations were reported for the first time in our study, as far as we are concerned, not even in Mozambique. Escobar and co-workers detected a non-synonym mutation (V494I) in two samples from Maputo [26]. Another study from Mozambique reported 4 polymorphisms (L619L, F656I, V666V, and G690G) with no mutations associated with resistance to the same drugs aforementioned [27]. A recent study with samples collected in 4 study settings from Mozambique (Montepuez, Moatize, Mopeia, Massinga) in 2018, in the context of the therapeutic efficacy, detected synonymous mutations in two samples [25]. In Uganda, 5 non-synonymous and 1 synonymous, no validated or associated to resistance mutation were reported by [11], while in Tanzania, 31 mutations in codon R561H were reported by [46], however, none of these mutation were associated with artemisinin resistance[45]. The same mutation was reported in Rwanda, with the addition of another mutation at codon P574L, in a therapeutic efficacy study in children under 6–59 months old in 2018 [22]. Although these results illustrate an emergence of resistance cases in Africa, the reported non-synonymous mutation from our study is not enough to raise alarm in these countries, as well as in Mozambique. Furthermore, the latest therapeutic efficacy study in Mozambique confirmed the susceptibility of P. falciparum to the drugs currently in use [15], however, continuous surveillance of ACTs resistance markers is urgently needed.
We predicted the functionality of the protein. Thus, the obtained SIFT scores (Fig. 1) for V454E, V494L, A578P, V581A, E509D, E455Q and L663I indicate that these mutations are tolerated or have non-deleterious effect on protein function, because they are equal or above the cut off value of 0.05 [31–33]. For the remaining non-synonymous mutations, SIFT scores predicted them to be deleterious, with a score of 0.00, bellow cut off value of 0.05 [31–33]. SIFT issued a warning stating that there is low confidence in this prediction, since these substitutions may have been predicted to be deleterious just because the sequences used were not diverse enough. According to Pauline C. Ng and Steven Henikoff, SIFT builds alignments with a median conservation value of 3.0. Usually, predictions based on sequence alignments with higher median conservation values are less diverse and will have a higher false positive error [31].
Our results showed that we do not have an emergency of artemisinin resistance cases, even though we observed novel non-synonymous mutations, highlighting the relevance of increasing the number of studies focused on the molecular surveillance of resistance markers to ACTs.