Background
The rise of drug-resistant parasites threatens to hamper malaria containment strategies. After the introduction of artemisinin-based combination therapy (ACT) in Jimma zone in 2006 [
1], a certain shift in the distribution of wild types and mutations of defined single nucleotide polymorphisms (SNPs), associated with anti-malarial drug resistance, was expected.
The
Plasmodium falciparum chloroquine resistance transporter (
pfcrt) gene on chromosome 7 encodes a transmembrane protein to be found in the digestive vacuole of the parasite and has primarily been associated with resistances to chloroquine (CQ). However, it also seems to influence artemisinin, quinine and amodiaquine susceptibility [
2‐
4]. The
P. falciparum multi-drug resistance protein 1 (
pfmdr 1) gene, located on chromosome 5, encodes a P-glycoprotein homologue 1 [
5,
6]. Point mutations in codon 86 and 184 seem to be involved in artemether-lumefantrine (AL) resistance in some [
7‐
10] but not all studies [
11]. Concerning CQ, the PfMDR 1 protein is believed to add a modulatory effect to resistance mechanisms without conferring actual resistance [
12]. An increased copy number has been linked to affect mainly mefloquine, but also artesunate, dihydro-artemisinin, halofantrine, quinine, and lumefantrine susceptibility [
13‐
15]. Increased copy numbers of this gene however, do appear to play a minor role on the African continent [
8,
9,
16].
The
P. falciparum SERCA-type ATPase 6, encoded by the
pfATP 6 gene, was suggested to be the cytosolic target structure of artemisinins [
17‐
19]. A variety of polymorphisms located in the
pfATP 6 gene have been reported [
20‐
23]. However, their actual importance concerning artimisinin resistance remains unclear. Recently, mutations in the
K13-propeller gene encoding the PF_1343700 kelch propeller domain have been proposed to be determinants for artemisinin resistance [
24]. Yet its actual function within the parasite organism is unknown and can only be assumed, as homologous proteins in other organism can be found participating in a wide variety of pathways [
24‐
26].
The
P. vivax multi-drug resistance protein1, encoded by the
pvmdr 1 gene, is the
pfmdr 1 orthologue in
P. vivax and is presumed to be connected to changes of CQ, amodiaquine and sulfadoxine-pyrimethamine treatment response. Specifically mutations in codon 976 and 1076 seem to be responsible for the aforementioned associations [
27‐
29].
In this present study, the prevalence of these molecular markers in southwestern Ethiopia collected in 2013 was compared to earlier published and partially unpublished data from 2004, 2006 and 2009 from this region [
30‐
32].
Discussion
In this study, molecular resistance patterns in Ethiopian
P. falciparum and
P. vivax isolates were investigated.
Pfcrt 76 T, associated mainly with CQ resistance, could be found almost as frequently as before introduction of ACT. Either continuous CQ use remains common in Jimma region, or the
pfcrt wild type provides no fitness advantage under AL treatment. As the wild type 76 K re-emerged under ACT and in absence of CQ in some countries [
39], continuous CQ treatment in Jimma region should be at least suspected. CQ is not recommended nor released in the health centres for
P. falciparum in Jimma region but available on the free market and recommended for
P. vivax treatment
.
As shown above, microscopic diagnosis was not always reliable; performance was similar in all health centres. Only in 203 (60.1 %) of all samples, the microscopy result matched the PCR result. A total of 16 samples microscopically diagnosed as vivax malaria could be proven to be
P. falciparum infections by PCR. Overall, 19.8 % microscopically positive samples were negative by PCR, 21 were microscopically categorized as vivax malaria and treated accordingly. Therefore,
P. falciparum parasites still have opportunities to get in contact with CQ. Also, self-treatment of the rural population in less severe cases seems to be still very common as CQ is cheap and easily available. The prevalence of
P. vivax in Jimma region is between 30-70 % [
40]. The performance data are consistent with a recent study from Ethiopia assessing malaria diagnostic capacities in Ethiopian health centres [
41]. A recent study from Mozambique showed a very similar changing pattern. The
pfmdr 1 allele N86 increased from 19.5 % in 2003–2005 to 73.2 % in 2010–2012 after introduction of AL [
42]. Interestingly, a recent study from southeast Ethiopia (Omo Nada, Bala Wajo, Arba Minch and Harar) showed different results [
43]. The authors declared a consequent absence of CQ use in this area. Prevalence of 76 T was 13.5 % (23/170) in the South and 32 % (8/25) in the East. Overall, the C72S mutation was observed only in 3.6 %.
The authors further stated that the CQ-sensitive CVMNK haplotype was found in 95.9 %, the mutant haplotype SVMNT in 4.1 % and CVIET was absent at codons 72–76. Sequencing of the mutant samples from Jimma showed in 100 % the CVIET haplotype in 2009 and 50 % CVIET / 50 % CVMNT in 2013, in accordance with other studies from Africa [
32]. The C72S mutation was absent as well as the sensitive CVMNK haplotype. However, a shift to CVMNT might be a possible interpretation. These regional differences are highly interesting and warrent further investigation.
The
Pfmdr 1 wild type 86 N was frequently detected in areas where AL is used. A selection for or re-introduction of the wild type is discussed for artemether as well as lumefantrine and might be the explanation for this drastic change [
7‐
10]. Since the introduction of AL in Jimma zone in 2006, the prevalence of the wild type remarkably increased (
P <0.001, data from 2006 compared to 2013 by Wilcoxon rank-sum test; Fig.
1). Comparing the
pfmdr1 184 haplotype with data from 2009, the prevalence of the mutation remained stable at above 98 %. These findings support the theory that the
pfmdr1 184 F is selected under drug pressure, especially under AL treatment [
9]. Unfortunately, no earlier data exists for
pfmdr 1 Y184F from this area. A recent study from southeast Ethiopia reported a prevalence of 9 % for N86 [
40]. This is similar to the data observed in this study. The prevalence of Y184F was 5 % only compared to 100 % in this study. Different selection mechanisms might be responsible. The diagnostic tool used in this study consisted of a real-time PCR with selective probes, sequencing was only performed in 10 % of the samples but the latter consistently confirmed the results of the PCR.
In 2009, all samples from Jimma showed only one
pfmdr copy. In the above-mentioned study from southeast Ethiopia, all samples presented also with only one copy [
43]. This is in accordance with other studies from Africa [
8,
9,
16].
The prevalence of the E431K mutation in the
pfATP 6 gene dropped from 58.3 % in 2006 [
31] to 18.8 % in 2013. Recently, E431K together with A623E has been described to reduce artimisinin susceptibility [
20]. The A623E mutation was extremely rare with 1/24 (4.2 %) in 2006 and 0/33 (0 %) in 2009. No samples with both mutations could be detected. The mutation was not assessed in 2013 but the decline of the E431K mutation contradicts selection of it under artemisinin pressure.
The new potential candidate referring resistance to artemisinins, the
K13 propeller gene, showed no mutations at codon 476. This mutation has been described by Ariey
et al. in highly resistant samples from Cambodia [
24]. A new mutation at codon N531I was found in only one isolate. Clinical effectiveness of AL was still 95 % in 2009. Clearance rates showed no significant prolongation, no resistance could be suspected then. More recent clinical data from this area were not available [
40].
Sequencing of the
pvmdr 1 gene revealed a high prevalence of the Y976F and F1076L mutations. This correlates well with the long lasting CQ use in this area. Treatment failures were reported in recent
P. vivax studies [
43‐
46]. Comparison of the Ethiopian
pvmdr 1 sequences to sequences from Brazil (AY571984.1), Cambodia (JQ925836.1), India (KC818412.1), Korea (GU476519.1, GU244390.1) Madagascar (EU683815.1) and Thailand (KC121338.1) showed highly conserved features without systematic geographical clustering.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made.
The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
AH carried out the molecular genetic studies, participated in the sequence alignment and drafted the manuscript. NA and AZ participated in the sample collection and coordination in Ethiopia. TE participated in design and coordination of the study. TL participated in design of the study and manuscript preparation. AW participated in sample collection and drafting of the manuscript. MP participated in coordination in Germany and drafting of the manuscript. NBR conceived of and designed the study, supervised the molecular studies and sequence alignment coordination and drafted the manuscript. All authors read and approved the final manuscript.