Background

Plasmodium falciparum malaria is the one of the major cause of morbidity and mortality in sub-Saharan Africa, including Angola, where three million clinical cases and 6000 deaths occurred in 2015. The deaths reported in Angola territory, accounting for 35 % in children under five years old and 25 % of the maternal deaths (Angolan National Malaria Control Programme, 2015).

In endemic malaria communities pregnant women and infants are more vulnerable to malaria episodes [1]. The World Health Organization strategy to protect mothers during their pregnancy and consequences in newborns such as low weight includes the implementation of intermittent preventive treatment in pregnant women (IPTp) [2]. This treatment was introduced in Angola in 2006, using the sulfadoxine–pyrimethamine (SP) at the second trimester of pregnancy. However, the emergence of P. falciparum resistances to both sulfadoxine and pyrimethamine can jeopardize these strategies.

Sulfadoxine–pyrimethamine resistant parasites are frequent in Southeast Asia and are becoming more and more prevalent in several East African countries [3]. SP acts by inhibiting the P. falciparum dihydrofolate reductase (dhfr) and dihydropteroate synthetase (dhps), two fundamental enzymes involved in the folate biosynthesis pathway [4]. Single nucleotide polymorphisms (SNPs) leading non-synonymous mutations in these two genes have been shown to be associated with pyrimethamine (dhfr) and sulfadoxine (dhps) resistances. The Angolan National Malaria Control Programme intends to introduce IPT in infants, as has already been done in other sub-Saharan Africa countries. To inform current policy makers when recommending the use of SP in IPT, dhfr and dhps molecular markers are used to differentiate high and low grade SP resistance areas and to define its geographical distribution [5].

Malaria is endemic throughout Angola, but the transmission patterns is heterogenous varying from intense transmission and hyperendemicity in northern, meso-endemic stable in the centre, to seasonal or epidemic unstable malaria in the southern part of the country [6]. No study for investigating pfdhfr and pfdhps mutations was performed in southern part of Angola where Lubango is located. The great majority of previous studies were conducted at the northern in Luanda, [7, 8]; Uige [9, 10], Cabinda, Kwanza Norte and Malanje [10], followed by those at the Central Angola in Benguela [11] and Huambo [10].

The objective of this study was to investigate the polymorphism at codons A16V/S, C50R, N51I, C59R, S108N, V140L and I164L in the pfdhfr gene and at codons S436A/F/C, A437G, K540E, A581G and A613T/S in the pfdhps gene, in order to provide baseline data regarding the proportion of P. falciparum pfdhfr and pfdhps mutations, before SP-IPTi introduction in Lubango, an Angolan malaria stable transmission region.

Methods

Study site, blood samples and DNA extraction

Blood samples were collected in 2011 from 52 patients presenting falciparum malaria at the Central Hospital in Lubango, located in Huila Province, South of Luanda, Angola capital. Lubango is located in southern Angola, with an area of 79,022 Km2, 1790 m above sea level, with an estimated population of 1,414,115. Malaria transmission occurs throughout the year, with peaks during rainy season, between March and May. Malaria is mesoendemic and P. falciparum is the predominant malaria species. Inclusion criteria comprised individuals with P. falciparum aging over 12 years and no evidence of complicated malaria. The Central Hospital is the main health facility in Lubango and it is a reference for malaria in the area. The malaria diagnosis was performed by thick blood smear and nested-PCR, as previous described [12]. All malaria cases were treated following the Angolan malaria treatment policy. After obtaining informed consent, venous blood collection was performed according to protocols approved by the National Institute of Public Health and the Ethical Committee of Angola and Fiocruz, Brazil (#372/07). The samples were cryopreserved and stored at −20 °C until DNA extraction. Genomic DNA was extracted from 1 mL whole blood using the QIAamp Midi columns (Qiagen), as described by the manufacturer.

Nested polymerase chain reaction (PCR) and electrophoresis

pfdhfr and pfdhps genes were amplified by nested-PCR approach using two gene-specific primers (external and internal) as already reported [13]. The nested-PCRs were performed to detect the presence of mutations at codons: A16V/S, C50R, N51I, C59R, S108N, V140L and I164L in the pfdhfr gene and at codons S436A/F/C, A437G, K540E, A581G and A613T/S in the pfdhps gene. The choice of these codons was based on the pioneer study of Pearson and colleagues [13].

DNA sequencing and SNP polymorphisms detection

After purification of the amplicons using the Wizard SV Gel and PCR Clean-Up System (Promega), PCR products were sequenced using Big Dye® Terminator Cycle Sequencing Ready Reaction version 3.1 (Applied Biosystems) and ABI PRISM 3730 DNA Analyzer [14] at the Genomic Platform/PDTIS/Fiocruz. The sequences of the amplicons were aligned with the wild-type 3D7 strain for pfdhfr (GenBank accession number XM_001351443) and for pfdhfr (GenBank accession number Z30654). The presence of SNPs was confirmed by reading both the forward and the reverse strands using the free software Bioedit Sequence Alignment Editor Version 7.2.5. Parasites with mixed alleles (in which both wild-type and mutant alleles were present) were considered mutants for estimation of the prevalence of the SNPs. Haplotypes for drug resistance markers were reconstructed from the full sequence presenting an unambiguous single allele signal at all positions. Statistical significance of differences between the haplotypes was assessed using Fisher’s tests and a p value <0.05 was considered significant.

Results

Among the 52 successfully sequenced samples for dhfr, 51 (98 %) showed non-synonymous mutations. The most frequent mutation was 108N (51/52; 98 %, 95 % CI) followed by 59R (33/52; 63 %, 95 % CI) and 51I (24/52; 46 %, 95 % CI). The 50R and 164L mutations were detected at the same frequencies (1/52; 2 %, 95 % CI) and no mutations were found at codons 16V/S and 140L (Table 1).

Table 1 Plasmodium falciparum dhfr and dhps mutated codons in parasites from Lubango, Angola
Full size table

Double mutants (51I/108N, 59R/108N) were observed in 39 isolates (76.5 %; 39/51) and triple mutants (51I/59R/108N, 50R/51I/108N, 59R/108N/164L) in ten (19.5 %; 10/51) P. falciparum samples. The 108N single mutant was found in only two (4 %; 2/51) isolates (Table 1). In samples presenting dhfr double mutation, the CNRN (59 + 108) allele was the most frequent (24/46 %) followed by CICN (51 + 108) (15/29 %). This contrasts with the frequency of the dhfr triple mutant alleles which were detected at lower rates: IRN (51 + 59 + 108), CNRNVL (59 + 108 + 164) and RICN (50 + 51 + 108) (8 (15 %), 1 (2 %) and 1 (2 %), respectively). The wild-type ACNCSVI was presented in only one isolate (Tables 2 and 3).

Table 2 Prevalence of P. falciparum dhfr and dhps mutants in parasites from Lubango, Angola
Full size table
Table 3 Deduced haplotype profiles for dhfr and dhps P. falciparum parasites from Lubango, Angola
Full size table

Of the 52 samples tested, 29 (56 %) were successfully amplified for the dhps gene. Among them 28 (97 %) presented a mutation at codons S436A, A437G and K540E. The 437G mutant allele was highly prevalent (97 %), while the 540E (7 %) and 436A (3 %) were observed in lower frequencies. No 581G and 613T/S alleles were observed (Table 1). The frequency of the single mutant SGKAA (437) (25/86 %) was higher than double mutants SGEAA (437 + 540) (2/7 %) or AGKAA (436 + 437) (1/3 %) (p < 0.05). The dhps wild-type SAKAA was rare and observed in only one (3 %) P. falciparum sample (Tables 2 and 3).

Combining pfdhfr and pfdhps alleles, not all isolates had an interpretable haplotype (Table 4). The PCR amplification failure rate of the dhps gene in Lubango samples might be somehow attributed to primer mismatched due to unknown polymorphisms in target sequences, because in all the 52 samples amplification was achieved when primers strictly developed to amplify conserved sequences for diagnosis purposes were tested [12]. The number of samples herein considered was based on amplification of both PCRs just to avoid the identification of incomplete haplotypes.

Table 4 Combined molecular dhfr and dhps gene profiles in 25 samples of P. falciparum from Lubango, Angola
Full size table

Two triple mutant haplotypes (double mutant in dhfr and single mutant in dhps) were observed: the ACICNVI/SGKAA in 14 (56 %) samples and the ACNRNVI/SGKAA in five (20 %) samples. One quadruple mutant haplotype was detected (ACIRNVI/SGKAA) in six (24 %) P. falciparum samples (Table 4). The high-grade SP resistant haplotypes (quintuple, sextuple and septuple pfdhfr/pfdhps) [3] were not seen among these isolates.

Discussion

At present, SP continues to be used in IPTp and IPTi strategies, which are likely to provide sub-optimal effect. Therefore, monitoring the spread of SP resistance using dhfr and dhps as molecular markers are strongly recommended [15]. This study was performed with P. falciparum samples from Lubango, an Angola region with unstable meso-endemic malaria transmission. The vast majority of tested isolates (98 %) presented at least one mutation in the gene encoding enzymes associated with pyrimethamine resistance.

High prevalence of 108N mutation (considered as the initial mutation occurring in dhfr multiple mutants) was detected, similar to observations already done in several Angolan regions [8, 10, 11] and in other African countries [16–26]. However, contrary to various Angolan localities, the 59R was the second more frequent dhfr mutation in Lubango, instead of the 51I [8–11]. In addition, it is worth to noting that the 164L pfdhfr SNP was also detected for the first time in Angola. This mutation is rare in Africa [5] but its presence in association with 108N + 51I and/or 59R [27] as well as with the dhfr + dhps quintuple mutant [5], confers high levels of in vivo and in vitro resistance that could render SP totally ineffective. The 164L mutation was also combined to cycloguanil resistance [28] and is commonly found in P. falciparum samples from Southwest Asia [29, 30] and South America [31]. The possible reason for the detection of mutant 164L in Lubango (north of Angola) and not in other correlate Angolan studies (North and Central Angola) is, probably, due to the greater proximity of Lubango with Zambia, where the presence of quintuple mutants is greater than 50 % [3].

From previous reports, it was known that the triple African mutant (N51I + C59R + S108N) shares a common ancestry with resistant dhfr from Southeast Asia [32, 33]; this mutant emerged in Asia and was introduced later in Africa, including Angola [19, 20, 33–36]. With respect to dhfr mutation 50R, another non-frequent dhfr mutation in Africa, it was also identified in this study. This low-grade pyrimethamine resistant mutant [27] was first described as specific of South American isolates [27] but, besides South America, it has already been described in Africa in isolates from Kenya [35], Central African Republic [37] and Angola [8].

In relation to dhps gene, the 437G mutation was the most common, reaching almost fixation, similarly to other Angolan localities [8, 10, 11] and other African countries [16, 20, 23, 26], especially in western and Central Africa.

The dhps 540E mutation, less commonly observed in western Africa, was also identified in this study, confirming previous data from Luanda [7, 8], Cabinda, Huige, Kwanza Norte, Malange and Huambo [10]. However, the spatial distribution of this allele in Angolan provinces seems heterogeneous and even in the same province its frequency can vary overtime [9, 10]. Thus, screening for dhps 540E, a surrogate for the quintuple mutant, is a priority in West Africa where it is comparatively rare [3].

The combination of 437G with the triple mutant 51I + 59R + 108N which is considered to be associated with SP treatment failure [38–40], was detected in 33 % of Lubango samples. However, in vitro studies showed that the single 437G mutant has a lesser degree of tolerance to sulfadoxine than the 437G + 540E double mutant. Thus, triple mutant 51I + 59R + 108N associated with single 437G mutant could have a less detrimental effect on SP-IPT than the associated 437G + 540E mutations—the sustained protective efficacy of SP-IPTi in western and Central Africa reinforce these points [41].

Despite the presence of parasites with triple dhfr mutant at positions 51I, 59R and 108N (15 %) and double dhps mutant at positions 437G + 540E (7 %), the quintuple mutant (51I + 59R + 108N + 437G + 540E), considered a significant predictor of SP treatment failure [42] was not detected in the studied isolates from Lubango. To date, this quintuple mutant was not observed in central and northern areas of Angola [7–11]. Besides Lubango, similar low frequencies of double 437G + 540E pfdhps mutant were described in Huambo (12.5 %) [10], Benguela (8 %) [10], Cabinda (5.3 %) [10], Malanje (3.1 %) [10], Luanda (3 %) [8], Kwanza Norte (2.1 %) [10] and Uige (1.9 %) [10], contrasting with the higher frequencies of this mutant detected in Eastern Africa: Kenya (86 %), Tanzania (90 %), Malawi (95 %) and Uganda (95 %). With the exception of Malanje (3.9 %), the frequencies of the triple 51I +59R + 108N dhfr mutant in Angolan provinces, including Lubango, were higher than the double 437G + 540E pfdhps mutant but, East African countries presented higher prevalence rates [43].

These findings are expected because East African countries presented a high rate of SP treatment failure of falciparum malaria together with the increased presence of the quintuple mutant considered the most specific molecular marker for SP [44, 45].

Besides quintuple mutants, the concept of “super resistant genotypes” is emerging which further raise the threshold of drug tolerance in parasites comprising additional mutations in combination with N51I + C59R + S108N and A437G + K540E [3]. The additional mutations are dhfr I164L (found in one sample; 2 %) and dhps A581G and A613T/S, which were not found in this study. Although, these mutations are rare, their impact on SP-IPTi or SP-IPTp efficacy is considered highly detrimental [5]. When the quadruple mutant N51I + C59R + S108N + A437G occur in the absence of 540E, these parasite populations are classified as partially resistant genotypes, and its resistance levels are not expected to be as high [3]. The efficacy of IPTp in western Africa, where N51I + C59R + S108N and A437G exist in combination with 581G but not with 540E supports the view that these parasites should be regarded differently [3]. In this case, molecular markers of super resistance may be an important warning about failing efficacy of IPTp.

Many studies suggested that IPTp with SP remains effective, or at least it is not associated with harm, in areas with high prevalence of P. falciparum quintuple mutant parasites [46, 47]. There is not enough evidence yet to establish a threshold prevalence of the classical dhfr + dfps quintuple mutant, or even of the dhps 581, dhps 540 or dhfr 164 point mutations above which there is a clear loss of IPTp-SP effectiveness. Therefore, the significance of the additional mutation at codon 540 needs further investigation.

Conclusion

In view of the data reported here and elsewhere, it was conclude that, although the number of samples available for the present study was limited, pfdhfr and pfdhps gene mutations in isolates from Lubango are suggestive of a low-grade SP resistance and their presences highlight the importance of molecular markers screening to monitor the evolution of P. falciparum SP resistance in Angola.