Molecular survey of pfmdr-1, pfcrt, and pfk13 gene mutations among patients returning from Plasmodium falciparum endemic areas to Turkey

Abstract

Background

In recent years, there has been an increasing trend in the number of imported Plasmodium falciparum cases in Turkey. To improve treatment success and to better understand malaria epidemiology among imported cases, it is necessary to determine anti-malarial drug resistance. This study aimed to survey polymorphisms of resistance genes in imported P. falciparum patients using archived thin smear preparations and EDTA blood samples.

Methods

A total of 100 imported P. falciparum patients admitted to Bakırköy Dr. Sadi Konuk Research and Training Hospital between 2017 and 2022 were included in this study. DNA extraction was performed using an archived slide and EDTA blood samples that were microscopically diagnosed. After confirming the samples by real-time PCR, the pfmdr1, pfcrt, and pfk13 genes were amplified and sequenced. Single nucleotide polymorphisms (SNPs) were screened using Geneious R9 software, with the reference P. falciparum clone 3D7 isolate.

Results

All studied samples were confirmed to be P. falciparum using real-time PCR. Nested PCR was conducted and the pfcrt (92 samples), pfmdr1 (91 samples), and pfk13 (93 samples) genes were successfully amplified. Sequence analysis revealed the highest mutation rate in the pfmdr1 (74.5%) gene, with the identification of five haplotypes: NYSND (wild-type, 23%), NFSND (56%), NYSDD (2.2%), NFSDD (15.4%), and YFSND (3.4%)]. The pfcrt mutation was identified in 11 samples (12.2%), whereas the pfk13 mutation was found in only two samples.

Conclusion

This study is the first molecular survey of anti-malarial drug resistance genes in Turkey. With the increasing number of imported Plasmodium malaria cases and recent reports of sporadic indigenous P. falciparum cases, malaria is becoming a growing concern in Turkey. Although molecular screening for resistance markers in P. falciparum malaria is not routinely conducted, the data from this study will enhance treatment success rates and contribute to global malaria elimination.

Background

With the increasing number of cases in recent years, malaria remains a substantial global health challenge, particularly in sub-Saharan Africa, Southeast Asia, and parts of Latin America. Despite advancements in prevention, diagnosis, and treatment, significant challenges persist in healthcare systems worldwide. The World Health Organization (WHO) estimates that between 2000 and 2021, there were 627,000 malaria fatalities and 247 million cases globally, with sub-Saharan Africa accounting for the majority of these cases [1]. Although there was a declining trend in malaria infections starting in 2000, a slight increase in the number of cases was observed between 2020 and 2021. Both the WHO and ECDC reported that case incidence and mortality rates increased in 2020, primarily due to disruptions in services during the COVID-19 pandemic [1, 2]. Turkey achieved malaria elimination in 2010, with zero indigenous malaria cases; however, both local and indigenous malaria cases have been reported recently [3, 4].

According to WHO, one reason for the increase in cases originating from Africa in 2020 was the emergence of Plasmodium strains resistant to anti-malarial treatments [5]. Since the first appearance of chloroquine resistance in the 1960s, various anti-malarial drugs have been used, leading to the emergence of drug resistance. This has contributed to treatment failure in recent years, resulting in inadequate treatment, relapse, and other complications. Recent evidence of partial artemisinin resistance in Africa has raised global concern. Therefore, WHO has identified studies to define the scope of resistance in terms of therapeutic efficacy and genotypic surveillance as an urgent priority.

Resistance in malaria parasites arises from spontaneous chromosomal point mutations or gene duplications. Chloroquine (CQ) mainly disrupts haemozoin formation in Plasmodium, leading to elimination the parasite due to the toxification of haem released from digested host haemoglobin [6]. Point mutations in the P. falciparum chloroquine resistance transporter (pfcrt) gene, particularly the K76T mutation, are associated with chloroquine resistance, not only in Africa but also globally [7]. CQ resistance (CQR) can occur either alone or in combination with another resistance gene, P. falciparum multidrug resistance 1 (pfmdr1). Studies on the pfmdr1 gene have identified five single nucleotide polymorphisms (SNPs) associated with resistance: N86Y, Y184F, S1034C, N1042D, and D1246Y. Polymorphisms in these codons are linked to anti-malarial resistance to drugs, such as quinine, chloroquine, mefloquine, lumefantrine, halofantrine, and artemisinin.

In recent years, the P. falciparum kelch propeller domain on chromosome 13 (pfk13) has been identified as a key factor in artemisinin resistance in numerous studies. Artemisinin, the primary drug used for malaria treatment, delays parasite clearance both in vivo and in vitro, underscoring its role in resistance. To date, more than 200 SNPs have been identified in the pfk13 gene of P. falciparum isolates, 10 SNPs confirmed to be associated with resistance. Mutations in the pfk13 gene are more commonly observed in Southeast Asia than in Africa, where they remain relatively rare [8, 9].

This is the first study to identify the molecular resistance profile of imported P. falciparum cases in Turkey. As recommended by the WHO, the surveillance of genetic mutations is crucial for effective malaria treatments and guiding global elimination strategies, given the serious implications of the spread of resistance. The aim of this study was to survey resistance-related SNPs in imported P. falciparum samples using archived thin smears and EDTA blood samples.

Methods

Sample collection

All samples were collected from patients admitted to the Bakırköy Dr. Sadi Konuk Training and Research Hospital (BEAH), according to the ethical approval given by the Clinical Research Ethics Committee with the protocol and decision number (2021/374, 2021/15/05). The samples used in this study were obtained from patients who had travelled from Africa to Turkey and were diagnosed with malaria through microscopic evaluation between 2017 and 2022.

DNA extraction from Giemsa-stained slides

Among the 100 P. falciparum samples included in this study, 45 were thin smears and 55 were EDTA whole blood tubes. The contents of the thin smear samples were scraped off with a sterile scalpel blade and prepared for DNA isolation and 200 µl of each EDTA sample was used for DNA isolation. DNA extraction was performed using the High Pure PCR Template Preparation Kit (Roche, France), according to the manufacturer's instructions.

PCR assay

Following DNA extraction from microscopically-investigated samples, real-time PCR was performed using species-specific primers (Fal-F primer/Falciprobe) to confirm P. falciparum infection. Extracted and measured DNA was used as a template, and a P. falciparum parasite isolate obtained from Ege University, Faculty of Medicine, Department of Parasitology, was used as a positive control [10]. The primers and probes used in this study were included in the standard operating procedures of University of Health Sciences laboratory for patients with malaria.

Amplification of resistance markers and SNP analysis

The polymorphisms of pfcrt, pfmdr1, and pfk13 were analysed via nested PCR amplification using previously reported primers. For the pfcrt gene, a region containing three loci (74, 75, and 76) was amplified [11]. For the pfmdr1 gene, three different primer sets were used to amplify five loci (86, 184, 1034, 1042, and 1246) [12]. Finally, pfk13 gene, which consist of eight loci (459, 469, 471, 561, 589, 602, 692, and 696), was amplified to identify resistance-related polymorphisms [13]. All PCR conditions for both the first and second rounds were followed as they were described in the original article, with no modifications [14]. The PCR products were visualized on a 1.2% agarose gel using 5 µl of 2nd round PCR products and observed bands were commercially sequenced (Macrogen, Seoul, Republic of Korea) bidirectionally to minimize or eliminate possible sequencing errors. Raw sequence data were analysed using Geneious R9 software, and the nucleotide and amino acid sequences of pfcrt, pfmdr1, and pfk13 of P. falciparum 3D7 strain (GenBank ID: Pf3D7_0523000, Pf3D7_0709000, Pf3D7_1343700) were used as references for alignment. SNP data were manually entered into Microsoft Excel and then transferred to SPSS 18 software (IBM) to perform the chi-square test (p < 0.01) and Fisher–Freeman–Halton test (p < 0.05).

Results

Demographic results

During the five-year period (2017–2022), 100 patients were included in this retrospective study, and demographic data, such as gender, nationality, and geographical origin of the patients were documented. Among all malaria patients, 79 were male, whereas 21 were female. Unfortunately, age was missing data and was not recorded for any of the patients. Most patients were non-Turkish (n = 86), and the reasons for travel were reported as study, tourism, and business (Table 1).

Table 1 Demographics of diagnosed imported P. falciparum patients

Imported country information was recorded for patients with malaria from West Africa, East Africa, and Central Africa. Although it is known that a group of patients (n = 22) travel from Africa, owing to the lack of travel history data and their uncertainty about where they were infected, the data for these patients were recorded as “no country info” (Fig. 1).

Fig. 1

Map showing the origin of imported malaria patients from Africa (map generated by MapChart, mapchart.net)

PCR assays

All samples included in this study were microscopically diagnosed as P. falciparum and identified at the species level by real-time PCR (Fig. 2). The total sensitivity and diagnostic accuracy of real-time PCR for smear samples were 82.2% (37/45) and 100% for EDTA blood samples (55/55). Eight smear samples that could not be detected as positive by real-time PCR were replaced with EDTA blood samples from the same patients.

Fig. 2

Real-time PCR amplification plot of P. falciparum positive slides

SNPs and haplotypes of resistance-related genes

Pfcrt: Second-round PCR products of 145 bp fragments were successfully sequenced in 92 samples. Sequence data covering 74–76 codons were analysed and three SNPs (M74I, N75E, and K76T) were identified in 11 patients.

pfmdr1: The pfmdr1 gene was successfully sequenced in 91 of 100 cases, and seven loci (86, 102, 182, 184, 1034, 1042, and 1246) were investigated for the presence of SNPs. Based on the pfmdr1 sequence data, the wild-type (NYSND) and four haplotypes (NFSND, NYSDD, NFSDD, and YFSND) were identified. The prevalence rates of mutated haplotypes NFSND, NYSDD, NFSDD, and YFSND were 56.2%, 2.2%, 15.7%, and 3.4%, respectively. The Y184F was the most frequently observed mutation, with a prevalence of 74.7% (68/91). The second most common mutation was N1042D, with a prevalence of 17.6% (16/91). The prevalence of N86Y mutation was 3.3% (3/91). In three cases from Nigeria, Chad, and Sierra Leone, a random base change was observed where thymine was replaced by cytosine in the “GGT” base, converting it to “GGC”. Since that replacement does not cause an amino acid change, G102G is a synonymous mutation. No SNPs were detected in codons 182, 1034, and 1246.

pfk13: A gene region containing eight codons (459, 469, 471, 561, 589, 602, 692, and 696) was sequenced in 93 cases, and the SNP causing E602D was noted in two samples.

Combined analysis of pfmdr1 and pfcrt: In 91 samples, both pfcrt and the pfmdr1 gene were successfully sequenced. While the prevalence of the Y184F mutation across Africa was 74.7%, it was higher (92.9%) in East African samples. The overall prevalence of Y184F in West Africa, Central Africa, and East Africa was significantly higher than that of pfcrt (IET) and N1042D (p < 0.01). The distributions of codons 74, 75, 76, 86, 102, 184, and 1042 were analysed separately by region, and no significant differences were found between the wild-type codons and SNP distributions (p > 0.05). Seven combined haplotypes were identified, namely, MNKNFN, MNKNFD, MNKYFN, IETNYD, IETNFN, IETNYN, and IETYFN. The dominant haplotypes were MNKNFN and MNKNFD with prevalence rates of 51.7% and 15.7%, respectively. The prevalence of the combination of the pfcrt mutation haplotype IET with one of the pfmdr1 mutation haplotypes (NYD, NFN, or YFN) was 7.9% (Table 2).

Table 2 Prevalence of haplotypes according to infection area

Within the scope of this study, 17 loci across three genes were examined, and based on the SNP findings, nine different haplotypes were observed (Table 3). All detected SNPs are presented in the supplementary file (Sup1).

Table 3 Observed haplotypes and SNPs among sequenced gene regions (green cells represent wild-type and red cell SNPs)

Discussion

Malaria is once again a rising concern in Turkey after its successful eradication in 2010. Several studies have reported the presence of both indigenous and imported P. falciparum. Additionally, a recent study reported P. falciparum/Plasmodium vivax mixed infections in three patients living in the same area who had no travel history [4, 15, 16]. Although molecular tests are not routinely used for malaria diagnosis, PCR-based assays are essential for applications such as vaccine research and anti-malarial drug resistance analysis. In the present study, the sensitivity of real-time PCR was 82.2% in blood smear samples and 100% in whole blood samples. The lower sensitivity in blood smear samples compared to whole blood samples may be due to the degradation of Plasmodium DNA in the smear samples over an extended storage period of up to four years, as well as potential DNA loss during the DNA isolation procedure. Similarly, a study conducted on whole blood samples reported a 97% detection rate of P. falciparum using real-time PCR [17]. Several studies have suggested using nested or real-time PCR to avoid false-negative results in the archived samples [18, 19].

The reduction in the use of chloroquine (CQ) and the adoption of artemether-lumefantrine (AL) as the first-line treatment has led to an increase in the prevalence of wild-type pfcrt K76 and pfmdr1 N86 and D1246 mutations in Africa [20, 21]. Findings of this study regarding to pfcrt and pfmdr1 are consistent with these observations. It has been reported that pfmdr1 N86Y and pfcrt K76T mutations make parasites more susceptible to lumefantrine and, consequently, to AL combination therapy [22, 23]. The high prevalence of wild-type K76 and N86 strains observed in this study suggests that increased use of AL may be contribute to the selection of these wild-type strains. Additionally, a decrease in these mutations has been associated with reduced anti-malarial efficacy of lumefantrine and, consequently, of AL [24, 25]. Isolates with the pfcrt IET haplotype exhibit increased resistance to CQ. The primary haplotype associated with CQ resistance is IET, and the N86Y mutation further supports CQ resistance [26, 27]. In this study, the IET (12.2%) and N86Y (3.3%) haplotypes were found at lower rates, similar to recent studies, which reinforces the increasing presence of CQ-sensitive P. falciparum in Africa [28,29,30,31]. The IET haplotype is also associated with increased resistance to AQ. Furthermore, studies have indicated that the N86Y and D1246Y mutations are linked to AQ resistance. The low detection rates of IET and N86Y, combined with the absence of D1246Y, suggest a high likelihood of success in malaria treatments involving amodiaquine combination therapies.

The pfmdr1 gene contributes to a multigenic resistance profile through mutations in different codons. In the present study, five previously reported loci (N86Y, Y184F, S1034C, N1042D, and D1246Y) were investigated, revealing five haplotypes (NYSND, NFSND, NYSDD, NFSDD, and YFSND) with varying prevalence rates. The susceptibility of P. falciparum to a broad range of anti-malarial drugs, such as quinine, mefloquine, amodiaquine, and artemisinin, is influenced by mutations in the pfcrt and pfmdr1 genes [32, 33]. Additionally, several variations in the pfcrt gene have recently been shown to confer resistance to piperaquine. Reports have also indicated that the Y184F mutation reduces sensitivity to lumefantrine [26, 27]. The widespread use of AL therapy has led to the selection of strains containing Y184F, which was the most frequently observed mutation in this study, consistent with findings from previous studies [34, 35]. The Y184F mutation does not directly cause phenotypic resistance, but alters drug transport kinetics, leading to decreased sensitivity. When combined with N86Y and/or pfcrt mutations, the Y184F mutation contributes to resistance. In this study, the IETNF (Y184F + pfcrt) haplotype was found in 4.5%, the MNKYF (Y184F + N86Y) haplotype was found in 2.2%, and the IETYF (Y184F + N86Y + pfcrt) haplotype was found in 1.1% of the samples. Therefore, while Y184F is prevalent, its presence alone does not directly affect anti-malarial drug resistance, which could pose a risk for recrudescence.

Recent resistance surveillance studies of African-origin P. falciparum cases have not detected the N1042D mutation [17, 27, 28]. However, in this study, the N1042D mutation, which contributes to quinine resistance, was found at a higher rate (17.4%) compared to other studies. Although quinine is not recommended as a first-line treatment for malaria, it is endorsed by the WHO for special circumstances such as pregnancy. Therefore, it is important to monitor the N1042D mutation associated with quinine resistance. In this study, the distribution of the pfcrt and pfmdr1 genotypes did not differ significantly among West Africa, Central Africa, and East Africa (p > 0.05). It is believed that the mutations detected are influenced more by the anti-malarial drugs used rather than by geographic regions.

Sequence data of the pfk13 gene from codon 440 to 700 was investigated. SNPs were detected at codon 602 in only two samples from Nigeria and Côte d'Ivoire. In a study evaluating the therapeutic efficacy of ACT in Tanzania, artemisinin resistance was not observed in a sample with E602D, and owing to the rarity of E602D, few studies have clearly established its relationship with artemisinin resistance [5, 36]. Another mutation, R561H, which causes delayed parasite clearance was first reported in a sample from Masaka, Rwanda [37]. According to the sequence data obtained in this study, no R561H mutations was observed in the studied samples. Results of this study also indicate that despite the widespread use of ACT, pfk13 mutations associated with artemisinin resistance have not yet been selected in Africa. One possible reason for the lower prevalence of pfk13 mutations in Africa compared to Asia might be the absence of mutations in DNA repair genes. Another study on resistant strains in Southeast Asia, where artemisinin resistance is more prevalent, reported differences in DNA repair genes that contribute to the development of artemisinin resistance [38]. These additional mutations can repair the DNA damage caused by artemisinin, helping mutant strains maintain their viability. Researchers have shown that when pfk13 mutants from Africa were compared with those from Asia, most African strains exhibited higher in vitro "fitness costs" (the energy required for the parasite to demonstrate resistance). This may contribute to the slower spread of artemisinin resistance in Africa [37]. Given the severe consequences of artemisinin resistance spreading in Africa, further studies are needed to explore the potential differences between the resistant strains in Asia and Africa.

Study limitations

There were several limitations of this study. In the scope of this study, certain demographic data for the patients, who presented to Dr. Sadi Konuk Training and Research Hospital could not be recorded. Information such as the patients’ ages and the diseases for which they were being monitored was not noted. Similarly, data regarding pregnancy status for female patients were unavailable. Detailed information on the treatment processes was also lacking, as patient follow-up occurred in a different unit. Additionally, biochemical parameters included in the study were analysed in a separate unit, so these data could not be presented within this study.

Conclusion

This is the first study to assess anti-malarial drug resistance in imported malaria cases in Turkey. This research provides essential baseline data for future studies in Turkey, a crossroad between Asia and Europe that hosts millions of refugees from Africa and Asia. Given the current data, molecular surveillance is crucial for both local and imported malaria cases. The findings will enhance the understanding of malaria epidemiology and are vital for updating treatment protocols.