Background
Malaria is a major global public health concern particularly in sub-Saharan Africa, with 219 million cases of malaria and approximately 435,000 deaths in 2017 [
1]. Most of the severe clinical cases and deaths were caused by
Plasmodium falciparum. Furthermore, pregnant women and children under 5 years old are the main victims of falciparum malaria. To alleviate the global malaria burden in a susceptible population, sulfadoxine–pyrimethamine (SP) is recommended by the World Health Organization (WHO) for use as intermittent preventive treatment in pregnant women (IPTp) and infants (IPTi) in malaria-endemic regions [
2].
Equatorial Guinea (EG) is a hyperendemic area of year-round malaria transmission [
3], and the population is more frequently exposed to episodes of malaria [
4]. Recent studies demonstrated
P. falciparum parasites are the predominant species in EG, leading to approximately 291,700 cases in 2016; 15% of the deaths from this species were in children under 5 years old [
5]. The authorities have deployed a series of measures that include effective anti-malarial drugs, vector control and case management for malaria control [
6]. In 2004, The Bioko Island Malaria Control Project (BIMCP) was initiated on Bioko Island [
7]. That project succeeded in reducing the infection rate, anaemia and child mortality [
6]. Subsequently, similar measures have been adopted and were applied on mainland EG by the Equatorial Guinea Malaria Control Initiative (EGMCI) in 2007 [
8]. In EG, SP has been used as a second-line treatment in cases of uncomplicated falciparum malaria for several decades. Furthermore, it was administered as the partner drug with artesunate as a first-line drug because of chloroquine treatment failure and as a malaria prophylaxis since 2004 [
9], which may have led to
P. falciparum isolates undergoing sustainable selection pressure. Soon afterwards, SP was replaced by artemisinin-based combination therapy (ACT) in response to widespread drug resistance in 2009, but it still remains the only choice for IPTp [
10]. Of even greater concern, SP resistance (SPR) had already evolved in most African countries before SP was implemented as the recommended treatment. To ensure the prophylactic efficacy of this approach and support the national anti-malarial policy, large-scale screening and surveillance of SP drug resistance is highly recommended [
11].
Targeting the
P. falciparum enzymes dihydropteroate synthase (DHPS) and dihydrofolate reductase (DHFR), SP acts as a synergistic inhibitor of folate in the parasite [
12,
13]. In vitro and in vivo studies have demonstrated that SPR is mainly conferred by amino acid point mutations at codons N51I, C59R, S108N, and I164L of
Pfdhfr and S436A, A437G, K540E, A581G, and A613S of
Pfdhps [
14]. These hotspot mutations are suggested to be gradually displayed with the increase of SPR [
15]. Many clinical failures have been reported after SP treatment was found in Africa [
16‐
18]. Thus, an urgent need exists to continue monitoring and assessing resistance in
P. falciparum populations when determining whether to administer this drug for prevention.
On Bioko Island, IPTp was introduced in 2004 [
7,
9], and the Ministry of Health has implemented the use of two doses of SP during pregnancy and antenatal care, starting from the second trimester and 1 month apart [
19]. An assessment of the prevalence of mutations in
P. falciparum genes related to SPR on Bioko Island is needed to provide complementary information for this preventive strategy. In the current study, an assessment of the prevalence of the
Pfdhfr and
Pfdhps gene mutations and haplotypes was conducted on
P. falciparum isolates collected from Bioko Island, EG.
Discussion
The rapid and widespread development of anti-malarial drug resistance is directly influencing and hindering the process of malaria control, prevention and elimination [
24]. Surveillance with molecular markers has allowed the early detection of drug resistance susceptibility and may provide fundamental information for drug policy [
25]. The current study displays the mutations and haplotypes of the
Pfdhfr and
Pfdhps genes from isolates collected from the general population on Bioko Island, thus allowing the degree of SPR in this malaria hotspot to be inferred.
The results demonstrate that
Pfdhfr polymorphism associated with SPR persists at high frequency. A high prevalence of the
Pfdhfr N51I mutation in 97.60% and the S108N mutation in 97.01% of the samples was found among the
P. falciparum population on Bioko Island (Table
1), and these mutations also had been found at a very high level (97.9 and 99.1%, respectively) in the Democratic Republic of Congo (DRC) in 2008 [
26]. For C59R, the level was significantly lower than for N51I and S108N, similar to observations in the mainland of EG [
4]. Like neighbouring countries,
Pfdhfr I164L, which is related to high-grade SPR, has been reported at low proportions (1.4%) in rural areas of the EG mainland [
4,
27]. Fortunately, this mutation was not found in any isolates within the study. Although the mutations of
Pfdhfr C50R and I164L are not found in the present data, the high prevalence of three well-characterized mutations in
Pfdhfr (N51I, C59R, S108N) indicate the
P. falciparum isolates from Bioko island display high pyrimethamine resistance that needs to be addressed by the EGMCI. For the
Pfdhfr haplotypes, 86.83% of the isolates carried the
Pfdhfr triple mutation (C
IRNI) (Table
2) and was reported in 80% of
P. falciparum infections in 2005 from the mainland of EG, 100% in 2005 in Cameroon [
28], and 72.4% in Gabon [
29]. This triple mutation is an important SPR indicator, but its detrimental effects may be largely compromised by an absence of the
Pfdhfr I164L mutation [
30,
31]. The frequency of the
Pfdhfr double mutant C
IC
NI was 5.99% (Table
2), and this genotype has a lesser degree of resistance compared with the triple mutation C
IRNI [
29]. For the dominant mutant haplotype C
IRNI (86.83%) and the double mutant haplotype C
IC
NI (5.99%), the results are consistent with previous studies in EG and Central Africa [
4,
10,
32,
33]. If the C
IRNI haplotype is found concurrently with the
Pfdhps mutations, it is associated with a high level of resistance [
34]. The reported prevalence of the
Pfdhfr triple mutation was also lower than those previously reported at the site where the proportion of the
Pfdhfr triple mutation reached a frequency of 97%. Only 1.2% of the isolates (2/167) were a pure
Pfdhfr wild type (CNCSI) (Table
2). The results indicate that almost all samples collected harbour pyrimethamine resistance.
Compared with the mutations of the
Pfdhfr gene, the mutations of the
Pfdhps gene exhibit a relatively low prevalence, except for the A437G mutation (90.51%, 143/158) (Table
1), which is also common in other EG regions and several African countries [
27,
31,
35]. This mutation has been reported to occupy the key position of the initial mutation of sulfadoxine resistance, and its resistance increases along with the augmentation of other mutations in
Pfdhps [
36]. Although the prevalence of S436A is significantly lower than that of A437G, it is higher than for other mutations, including K540E, A581G and A613S. In Central Africa, the
Pfdhps K540E mutation was less prevalent, which was also confirmed in this study (5.06%, 8/158) (Table
1). This mutation is more common in East Africa, particularly in Tanzania [
37] and Uganda [
17]. The WHO has recommended that IPT with SP should be abandoned in areas where the K540E mutation has been detected at > 95% and
Pfdhps the A581G mutations are detected at > 10% because it could be ineffective [
11]. Fortunately, only 5.06% (5/158) of the isolates showed the
Pfdhps K540E mutation, and 0.63% (1/158) of the isolates harboured the A581G mutation in current survey (Table
1). The relatively low prevalence of these mutations suggests that IPT-SP can possibly be efficacious on Bioko Island, EG. The A613S mutations were detected in 3.16% (5/158) of the isolates, which is consistent with reports in Central African countries, including the DRC [
27] and Cameroon [
29]. For the
Pfdhps haplotype, the single-mutant S
GKAA haplotype predominates in our results (62.66%) (Table
2), similar to observations made in Gabon [
38] and the DRC [
39].
AGKAA is present in 10.76% of the isolates (Table
2), and an increased trend was detected in Gabon between 2013 and 2014 [
38]. Parasites with double- and triple-mutant
Pfdhps haplotypes were observed at a low frequency (Table
2), suggesting a low tendency in the emergence and development of the sulfadoxine resistance alleles.
The combination of the
Pfdhfr and
Pfdhps mutant alleles generated 12 different haplotypes in the present survey (Table
3). Only one wild-type haplotype (CNCSI-SAKAA) was found in this study (Table
3). The quadruple mutant (C
IRNI-S
GKAA) was predominant, with a prevalence of 65.38% (Table
3), which is higher than reports from mainland EG (54%) [
4]. The saturation of the
Pfdhfr triple mutants could further induce the
Pfdhps mutants, and thus, the presence of quadruple mutants (C
IRNI-S
GKAA) was common [
40]. Although quintuple mutant genotypes (C
IRNI-S
GEAA) are highly linked to SP failure [
34], this mutant was detected at a rate of 4.62% (Table
3). WHO recommends surveillancefor this genotype and inhibition of IPTp-SP when the prevalence of this quintuple mutant exceeds 50% [
31]. To date, this quintuple mutant is less than 10% in other areas of EG [
4,
10]. Previous in vitro studies demonstrated that the quadruple mutant (C
IRNI-S
GKAA) has a less deleterious effect on SP-IPT than the quintuple mutant genotypes (C
IRNI-S
GEAA) [
41]. Notably, the ‘super resistant’ alleles (C
IRNI-S
GEGA) may render SP ineffective [
42], but these were detected in only one isolate. Although this occurrence is low, sustainable monitoring for SPR and avoiding the growth of super resistance alleles are still critical.
Although the LD analysis of the SNPs between the
Pfdhfr and
Pfdhps genes showed a strong linkage between N51I and A437G, those main SNPs of the
Pfdhfr and
Pfdhps genes form two independent LD blocks, respectively. These results indicate that the mutations located in the
Pfdhfr and
Pfdhps genes have relative independence. However, combined chemotherapy will likely lead to the occurrence and progress of resistance gene mutations even though the
Pfdhfr and
Pfdhps genes are located on different chromosomes [
40]. For the
Pfdhfr gene, T152A, T175C, and G323A develop as a block. When distributed in the
Pfdhfr gene, these SNPs exhibit strong linkage, particularly of N51I and S108N (D’: 0.71–1, P < 0.05). For the
Pfdhps gene, the T1482G, C1486G, A1794G, and G2013T were found in an LD block. Although the SNPs in
Pfdhps gene show weak linkage and no significant differences (P > 0.05), strong linkages were also commonly detected from S436A and other mutations, including A437G, K540E and A613S. Notably, the study had weaknesses, including the small sample size and the lack of full-length DNA sequences for the
Pfdhfr and
Pfdhps genes. In the present study, a 594-bp fragment of the
Pfdhfr and a 711-bp fragment of the
Pfdhps gene were amplified, based on previous study [
21]. The sequences from these two fragments provide only limited information for LD analysis. Thus, the complete nucleotide sequences from the
Pfdhfr and
Pfdhps genes and the microsatellite loci flanking these genes [
43] need to be amplified and genotyped in further study. Genetic diversity information and differentiation data from microsatellite loci flanking the
Pfdhfr and
Pfdhps genes will demonstrate whether the
P. falciparum isolates have ever undergone selection in response to SP and may provide valuable information to solve anti-malarial drug resistance, particularly SPR.