Molecular surveillance of Pfcrt and k13 propeller polymorphisms of imported Plasmodium falciparum cases to Zhejiang Province, China between 2016 and 2018

Malaria is a significant public health problem because of its worldwide distribution and high mortality. It was estimated that there were 219 million cases of malaria globally in 2017, mostly in 15 sub-Saharan African countries and India, representing approximately 80% of the global malaria burden [1]. Of the five malaria species, Plasmodium falciparum caused the most malaria incidence worldwide, accounting for 99.7% of estimated malaria cases in the World Health Organization (WHO) African region and 62.8% in Southeast Asia in 2017, and is the causative agent of the most severe forms of the disease [1].

Effective treatment is critical. Since the 1940s, multiple anti-malarial drugs have been developed and used to treat malaria parasites, including chloroquine (CQ), mefloquine, quinine, pyrimethamine, and sulfadoxine. However, the widespread use of these drugs promotes drug resistance, especially chloroquine resistance (CQR). Resistance to CQ occurred in Southeast Asia, South America, and the Western Pacific region in the late 1950s and rapidly spread to malaria-endemic areas, including Africa [2, 3]. Mutations in the chloroquine resistance transporter (Pfcrt) located on the P. falciparum digestive vacuole membrane were responsible for CQ treatment failure [4, 5]. Amino acid polymorphisms at PfCRT amino acid residues 72–76 were observed in CQR field isolates, whereas CVMNK haplotypes at PfCRT residues 72–76 were regarded as chloroquine sensitive (CQS) [6, 7]. Other studies revealed that Pfcrt K76T variants could affect parasite fitness, increase the rate of gametocyte production, and alter the susceptibility to artemisinin-based combination therapy (ACT) [3, 8‐10]. These results highlight the need to monitor the molecular evolution of Pfcrt.

In view of P. falciparum multidrug resistance, the WHO recommended ACT as the first-line treatment for uncomplicated P. falciparum malaria in 2006 [11]. However, the detection of artemisinin-resistant P. falciparum in western Cambodia and the border between Cambodia and Thailand in 2008 was a drawback to malaria elimination [12, 13]. Over 200 nonsynonymous P. falciparum Kelch13 (k13) mutations have been reported to date, of which nine variants (F446I, N458Y, M476I, Y493H, R539T, I543T, P553L, R561H, and C580Y) were correlated with slow parasite clearance and reduced in vitro drug sensitivity, and over 20 k13 mutations are considered candidates or associated markers [14]. k13 mutations were detected predominantly in the Greater Mekong subregion(GMS) [15]. k13 mutations are rare in Africa, and their profile is highly heterogeneous [16]. The prevalence of nonsynonymous k13 mutations is low in approximately 50% of African countries [15]. Nevertheless, the increase in drug resistance in Africa could hamper malaria control considering its high morbidity and mortality. Therefore, monitoring mutations associated with artemisinin resistance via delayed parasite clearance globally, but especially in Africa, is critical.

Zhejiang province, located in eastern China, was considered malaria-free in 2018. No indigenous malaria infections have been reported in Zhejiang province since 2011. However, approximately 200 cases are imported every year, especially P. falciparum malaria from Africa. In this study, samples were collected from P. falciparum cases imported into Zhejiang Province, China, between 2016 and 2018, and molecular surveillance of Pfcrt and k13 was performed to determine the emergence and spread of drug resistance in the countries of origin.

As one of the most important approaches to combat malaria, chemotherapy is paramount to treat Plasmodium and interrupt malaria transmission. However, effective anti-malarial drug policies were followed by drug resistance, which temporarily interrupted malaria elimination campaigns. Drug resistance has posed a significant threat to global malaria control strategies and raised international concern, especially in areas most strongly affected by the disease [19]. Most malaria cases originated in the WHO African region, accounting for approximately 92% of all cases and 93% deaths [1]. Four countries in Africa represented almost 50% of all malaria cases worldwide: Nigeria (25%), The Democratic Republic of Congo (11%), Mozambique (5%), and Uganda (4%) [1]. Furthermore, African countries spread the infection to other countries, including France (> 2500), China (> 2000), UK (approximately 1500), and Germany (> 500) in 2017 [1, 20]. Zhejiang province, located in southeast China, had several cases of P. falciparum malaria imported from Africa in recent years [21, 22]. The current study evaluated drug resistance markers of P. falciparum infections predominantly from Africa using molecular assays to advance drug policies against malaria.

Chloroquine, an easy-to-use and affordable first-line antimalarial agent, is comprehensively used globally to treat P. falciparum and Plasmodium vivax. Nevertheless, this drug was withdrawn from most endemic countries due to the high levels of resistance, which has resulted in a two- or threefold increase in malaria deaths and hospital admission for severe malaria in various African countries [19, 23]. Over ten multi-mutations sharing the common K76T substitution have been detected and are associated with CQR in field and laboratory strains of P. falciparum [4, 24, 25]. Five major haplotypes at Pfcrt residues 72–76 (CVIET, SVMNT, SVIET, CVMNT, and CVTNT) were related to CQR, and CVIET and SVMNT were regarded as the most resistant haplotypes [7, 26‐28]. CVIET is predominant in Africa, whereas SVMNT is more common in South America [27]. Our study confirmed that CVIET was the most common mutation type in infections imported from Africa, similar to previous studies [26, 28]. The high frequency of SVMNT in Papua New Guinea was also consistent with other studies [29]. It was demonstrated that the spatial distribution of mutant alleles was mainly related to local drug policy. The increase in the prevalence of SVMNT isolates in Tanzania from 0% (0/156) in 2003 to 3.68% (6/163) in 2004 was due to selective pressure for amodiaquine resistance [27]. Therefore, it was postulated that CVIET might be displaced by SVMNT in African regions where amodiaquine was increasingly used [27, 30]. However, our study refuted this hypothesis by showing the absence of SVMNT double mutants in Tanzania and other African countries. The disagreement in the results may be due to differences in sample size and survey sites. The present results indicate that parasites with CVIET are still a major threat in Africa.

With the cessation of CQ use between 1998 and 2008, parasite populations with wild-type CVMNK returned progressively, demonstrated by the increased frequency of this haplotype in Africa [30‐33]. For instance, the frequency of a CQS genotype increased from 28.0% in 2003 to 53.7% in 2012 after disuse of CQ for 9 years in Cameroon [34]. This hypothesis is supported in our study by the consistent detection of CVMNK (72.61%) in imported malaria cases and by another study wherein most P. falciparum isolates from Africa harbored wild-type alleles [33]. A possible explanation is that CVIET mutants were not fit enough to evolve into wide-type strains without selective pressure [30]. These results help understand the dynamics of major Pfcrt haplotypes in Africa and enrich the evidence for drug policy.

K13 propeller polymorphisms in P. falciparum have been widely studied since it was first described in GMS [12]. Molecular markers for k13 have been identified and can help elucidate artemisinin delayed parasite clearance [15]. In this study, 19 alleles were identified, and eight (P441S, D464E, K503E, R561H, A578S, R622I, V650F, and N694K) were nonsynonymous. Similarly with previous literatures, the propeller domain of the k13 gene showed a limited diversity of alleles in Africa [18, 35, 36]. Of note is that one of the validated k13 resistance mutations—R561H—was detected in two patients, one each from Rwanda and Myanmar. R561H in Myanmar and western Thailand, although not the predominantly popular, was more prevalent than that in Africa [33, 37, 38]. The only infected patient from Myanmar was positive for R561H. Previous studies found the R561H allele in Congo. However, R561H might be the first time found from Rwanda [36, 37, 39]. In addition, mutation A578S identified in one isolate from Cameroon, was also found in other African countries, including Gabon, Uganda, Mali, Kenya and DR Congo [18, 39‐41], but it was not associated with clinical or in vitro resistance to artemisinin according to previous studies [37, 42]. The study also found the uncommon k13 mutation R622I in two cases, one in Mozambique and one in Somalia, and this mutation was initially reported in northwest Ethiopia at the border with Sudan [43]. A previous study showed that, of a total of over 14,000 screened samples from 59 countries in Africa, Asia, Oceania, and South America, only one sample from Zambia was positive for R622I [37]. Interestingly, one out of three patients from Ethiopia bearing R622I mutation showed day-3 positivity in Giemsa-stained smears [43]. Further clinical studies are required to investigate the role of R622I in acquired resistance to artemisinin.

In Nigeria, where nearly 25% of the samples collected in this study originated, seven samples exhibited six mutant types (two nonsynonymous types comprising P441S and V650F, and four nonsynonymous types comprising C469C, V589V, N664N, and A676A). Similarly, no validated or candidate amino acid mutations were detected in southwestern Nigeria according to recent studies [35, 44]. However, a low prevalence of single nucleotide polymorphisms (G665C, V666V, P553P, V510V, A578S, D464N, and Q613H) were also reported, although they differed from the results obtained in the current study [35, 44]. In general, these results indicate that the profiles of the molecular markers conferring artemisinin delayed clearance were still optimistic with respect to the public health situation in terms of malaria in Nigeria.

The molecular genetic analyses conducted in this study showed that the haplotype diversity of the samples imported Africa was relatively low. Similarly, haplotype diversity of P. falciparum isolates in Congo, Ghana, Kenya and Tanzania in a previous study was 0.067, 0.123, 0.066 and 0.056, respectively [45]. Also, southwestern Nigeria reported Hd 0.080–0.157 in two states [35]. However, another research reported Hd 0.74 from African isolates [46]. The difference might result from heterogeneity of spatial distribution of samples. The current study also showed that the sequence diversity varied among the isolates from West, South, East, and Central Africa. The isolates from East Africa had the highest genetic diversity, whereas those from Central Africa had the lowest diversity. The discrepancy might have been related to the unequal sample sizes in different areas. Further investigation is required. Regarding the neutrality test, there was no consistent results. Tajima’s D value in this study was found to be negative and statistically significantly in terms of the total samples from Africa. Another research also reported similar trends from African isolates [46]. Nevertheless, negative Tajima’s D value without statistical significance was observed in Nigeria which indicated k13 gene evolved under neutral model [35].

In conclusion, this study determined the frequency and spatial distribution of Pfcrt and k13 mutations associated with drug resistance in P. falciparum malaria cases imported into Zhejiang province. In Africa, wild-type Pfcrt was predominant, but detection of k13 mutants was limited by the absence of genetically validated molecular markers of artemisinin delayed parasite clearance, except one case from Rwanda harbouring the R561H mutation. Both Pfcrt and k13 mutations were detected in patients from Southeast Asia and Oceania. These results may guide efforts to make more rational and targeted drug policies to eliminate resistant malaria in the study region and demonstrated that molecular markers are an effective and convenient tool to improve the surveillance of drug resistance.