The evolution of pyrimethamine resistant dhfr in Plasmodium falciparum of south-eastern Tanzania: comparing selection under SP alone vs SP+artesunate combination
© Malisa et al; licensee BioMed Central Ltd. 2011
Received: 20 August 2011
Accepted: 26 October 2011
Published: 26 October 2011
Sulphadoxine-pyrimethamine (SP) resistance is now widespread throughout east and southern Africa and artemisinin compounds in combination with synthetic drugs (ACT) are recommended as replacement treatments by the World Health Organization (WHO). As well as high cure rates, ACT has been shown to slow the development of resistance to the partner drug in areas of low to moderate transmission. This study looked for evidence of protection of the partner drug in a high transmission African context. The evaluation was part of large combination therapy pilot implementation programme in Tanzania, the Interdisciplinary Monitoring Programme for Antimalarial Combination Therapy (IMPACT-TZ)
The growth of resistant dhfr in a parasite population where SP Monotherapy was the first-line treatment was measured for four years (2002-2006), and compared with the development of resistant dhfr in a neighbouring population where SP + artesunate (SP+AS) was used as the first-line treatment during the same interval. The effect of the differing treatment regimes on the emergence of resistance was addressed in three ways. First, by looking at the rate of increase in frequency of pre-existing mutant dhfr alleles under monotherapy and combination therapy. Second, by examining whether de-novo mutant alleles emerged under either treatment. Finally, by measuring diversity at three dhfr flanking microsatellite loci upstream of the dhfr gene.
The reduction in SP selection pressure resulting from the adoption of ACT slowed the rate of increase in the frequency of the triple mutant resistant dhfr allele. Comparing between the two populations, the higher levels of genetic diversity in sequence flanking the dhfr triple mutant allele in the population where the ACT regimen had been used indicates the reduction in SP selection pressure arising from combination therapy.
The study demonstrated that, alleles containing two mutations at the dhfr have arisen at least four times independently while those containing triple mutant dhfr arose only once, and were found carrying a single unique Asian-type flanking sequence, which apparently drives the spread of pyrimethamine resistance associated dhfr alleles in east Africa. SP+AS is not recommended for use in areas where SP cure rates are less than 80% but this study reports an observed principle of combination protection from an area where pyrimethamine resistance was already high.
KeywordsEvolution pyrimethamine resistance microsatellites combination therapy
Artemisinin-based combination therapy (ACT) is the World Health Organization (WHO) recommended anti-malarial in areas where SP failure prompts replacement. Already, Tanzania has replaced SP with a combination of artemether and lumefantrine (Coartem®). However, SP remains the only option for intermittent treatment of malaria during pregnancy and an important drug in the intermittent treatment of infant. Therefore, priority remains for continuing surveillance and monitoring of genetic changes of SP resistance in the population.
SP kills parasites by inhibiting folate biosynthetic pathway [1–3]. Pyrimethamine competitively inhibits the enzyme dihydrofolate reductase, DHFR  while sulphadoxine does the same to inhibit the enzyme dihydropteorate synthase, DHPS [2, 3]. Resistance to SP is associated with point mutations in the genes coding for the two enzymes, dhfr and dhps [4–8]. Single mutation at codon 108 (S108T) confer low level of insensitivity to the drug . Additional mutations increases the insensitivity of the parasite to the drug; the double mutation (N51I+S108N) or (C59R+S108N) confers intermediate resistance [5, 9] increasing to triple mutations (N51I+C59R+S108N). An additional mutation at codon 164 (I164L) leading to quadruple mutant, cause highest level of resistance to SP. It is now known that, quadruple mutation exist in Africa in a very low frequency [10-13, this study], although recent study in Uganda  indicates its frequency is increasing.
There is currently conflicting information about how SP resistant alleles originate and spread across populations [13–20]. Early studies suggest that, the dhfr triple mutant allele is monophyletic in origin indicating it has arisen once  and subsequent studies reported the lineage had arisen in south-east Asia . Although many studies that followed confirmed this observation [14, 17–20], a study in an area of high malaria transmission in western Kenya identified presence of multiple lineages of the dhfr triple mutant allele , and concluded that, local evolutionary history as well as dispersal by gene flow are equally important in the establishment of resistance in populations.
This study has investigated the rate of selection and evolution of dhfr resistant alleles in two populations with contrasting anti-malarial therapies, Kilombero-Ulanga under SP monotherapy and Rufiji under SP+AS. By using three sets of microsatellite loci flanking the dhfr gene, the study investigated the evolutionary origin of resistant dhfr alleles and compares the rate of underlying genetic exchange between the two populations.
Study area, subjects and samples
Samples were collected through three community cross sectional surveys conducted during July-September in 2001, 2002 and 2006 in three rural districts of south-eastern Tanzania; Rufiji (Population = 203,000), Kilombero (Popupulation. = 322,000) and Ulanga (Population = 194,000). The surveys were part of large combination therapy pilot implementation programme in Tanzania, the Interdisciplinary Monitoring Programme for Antimalarial Combination Therapy (IMPACT-TZ). IMPACT-Tanzania is a multiyear implementation research evaluation that rests on a collaborative platform incorporating the US Center for Disease Control and Prevention (CDC), the Ifakara Health Institute (IHI), London School of Hygiene and Tropical Medicine (LSHTM), and the Ministry of Health and Social Welfare including its National Malaria Control Programme, the Tanzania Essential Health Interventions Project and the Council Health Management Teams of Rufiji, Kilombero, Ulanga, Morogoro and Mvomero Districts. For the purpose of the study, Kilombero and Ulanga Districts were treated as a single district because population movement between these two districts is high and the study population spans the border region. Plasmodium falciparum malaria transmission in the study area was intense (with an estimated entomological inoculation rate of 367 infectious bites per person per year  and perennial with some seasonal fluctuation (however, recent studies indicate decrease in malaria transmission in all endemic areas but new entomological inoculation rate for the studied area has not yet been re-determined). The survey conducted in Jul-Sept 2001 coincided with the implementation of new national policy replacing CQ with SP as the recommended first-line treatment of uncomplicated malaria. In 2002 the timing of the survey coincided with the implementation of SP+AS combination in Rufiji. While the national policy change to SP in 2001 was implemented nationwide, the change from SP to SP+AS combination in 2002 was applied in Rufiji district only.
A total of 22,696 adults and children belonging to randomly selected households participated in the study (6,482, 7,447 and 8,767 in 2001, 2002 and 2006 survey, respectively), and a finger-prick blood sample for blood slide and filter paper bloodspot were collected from each individual. The filter paper bloodspots were air-dried and stored at room temperature in self-sealing plastic bags with desiccant and stored at room temperature for future use in genetic testing. All blood slide samples were screened by microscopy for P. falciparum positivity. Bloodspots from microscopically positive subjects were selected for molecular genotyping.
Scientific and ethical clearance was granted from the Medical Research Council of the National Institute for Medical Research in Tanzania, the Centre for Disease Control and Prevention, USA, and the London School of Hygiene and Tropical Medicine, UK. Consent was obtained from all individuals or their guardians before collection of samples.
The DNA was extracted from bloodspots dried on filter papers. A sector of the dried blood spot filter paper was excised using a sterile blade or scissors, and soaked in a 1 ml, 0.5% saponin-1x phosphate buffered saline (PBS) overnight in a 96-deepwell plate. The segment was then washed twice in 1 ml of 1x PBS and finally, was boiled for 8 min in 100 ul PCR quality water with 50 ul 20% chelex suspension (pH 9.5).
Nested PCR was used to amplify a 594 bp fragment of dhfr containing the sequence where mutations are found. Primer sequences and PCR reaction conditions were previously described in  PCR was performed in 96 well plates with 25-μl PCR reaction volumes containing final concentrations of 0.25 μM oligonucleotide primers, 2 mM MgCl2, 250 μM each deoxyribonucleotide triphosphate (dNTPs), and 1x Taq polymerase. One microlitre (1 μl) of DNA template was used in the outer (primary) PCR reaction mixture. The outer dhfr PCR products were diluted three fold before a 1 ul was introduced into the inner PCR reaction mixtures.
Molecular genotyping of point mutations by Sequence Specific Oligonucleotide Probing (SSOP)
The nucleotide and amino acid substitutions at dhfr gene screened for by PCR-SSOP
Ser (S) AGC
Asn (N) AAC
PCR products were spotted in a 12 by 8-grid and cross linked onto nylon membranes and probed for sequence polymorphisms by hybridisation to specific oligonucleotide probes described previously . The probed blots were visualized using ECF substrate and detection using a phosphorimager (STORM). Inspection of the phophorimager output was recorded through viewing of digitally captured images of chemi-fluorescent signal.
The stringency and specificity of the hybridisation process was confirmed by inspection of a series of four controls of known single genotype variant sequence. All blots with non-specifically bound probes were stripped and re-probed. A SNP was considered to be present in the PCR product when the intensity of signal was higher than that of the background. The blots were scored independently by two people and the results reconciled.
The study aimed to establish the relative abundance of different point mutation haplotypes at dhfr. Since bloodstage P. falciparum is haploid this is very straightforward when an infection consists of a single genotype because only one form of sequence is seen at every SNP locus. When infections are composed of multiple genotypes a mixture of different sequence variants occur making the inference of point mutation haplotypes within that infection impossible.
Proportion of samples that were P.falciparum positive, PCR amplified dhfr, and single/majority dhfr
P. falciparum positive
PCR amplified dhfr
Single or majority dhfr
In order to analyse microsatellite immediately flanking sequence polymorphism on chromosomes carrying specific allelic forms of dhfr, the study used the genotype data obtained by PCR-SSOP method described above to identify and locate DNA sample bearing that particular allelic form. For the purpose of looking on these flanking sequence polymorphisms, the study deliberately selected a subset of samples that were unmixed at any polymorphic loci at dhfr (i. e. samples in which only single haplotype were obtained). Nonetheless, two or more alleles were still detected in some of the samples and such samples were classified as polyclonal and hence were excluded from haplotype construction and any further analysis. Microsatellite sequence located 0.3 kb, 4.4 kb, and 5.3 kb upstream of codon 108 of the dhfr-ts gene, which is on chromosome 4, were analysed. A semi nested PCR system employing the primer sequences and PCR cycling parameters described in  was used. A total of 728 samples comprising of 126 dhfr wild type (N51+C59+S108), 346 double mutant dhfr (N51I+S108N or C59R+ S108N) and 256 triple mutant dhfr (N51I+ C59R+ S108N) were selected from 2001, 2002 and 2006 genotypic data. Microsatellite PCR samples were diluted 1:100 and run with LIZ-500 size standard on an ABi 3730 genetic analyser (Applied Biosystems, Warrington, Cheshire, UK). Allelic sizes were separated on an Applied Biosystems ABi 3730 capillary sequencer and were scored using Gene Mapper software version 3.7 (Applied Biosystems, Warrington, Cheshire, UK).
Statistical comparison of allele frequencies at dhfr was carried out using Fishers exact test in STATA version 9.2. The calculation of binomial exact 95% confidence intervals was carried out using STATA version 9.2 . Gene diversity values was calculated as He = [n/(n-1)][1- ΣPi2] where n is the number of samples and Pi is the frequency of the ith allele.
Of 22,696 people sampled, 4,778 were found to be infected with P. falciparum. DNA was extracted from the 4,778 P. falciparum positive bloodspots and PCR amplification of dhfr performed once, giving a combined rate of PCR amplification success of 74% (Table 2). The amplified products were screened for all the variant sequences described in Table 1. Out of the 3,521 isolates which amplified successfully for dhfr, 2,565 (80%) were single or majority genotype infections and the point mutation haplotypes could easily be resolved.
In 2001, the frequency of the sensitive dhfr haplotype (N51+C59+S108) was significantly higher in Kilombero-Ulanga (48%) than Rufiji (30%) (P ≤ 0.0001). As the triple mutant dhfr allele increased in both districts, the sensitive dhfr allele was displaced proportionately reducing its frequency to (6%) in Kilombero-Uanga and (4%) in Rufiji (P ≥ 0.05) by 2006. The N51I+S108N and C59R+S108N double mutant dhfr alleles persisted throughout unchanged with an average frequency of roughly 10% at both populations, indicating they are independent of drug pressure (Figure 1).
The evolutionary origins of dhfr haplotypes was assessed by genotyping 728 isolates (CNCS = 126, CNRN + CICN = 346, and CIRN = 256) of P. falciparum carrying resistant or the sensitive dhfr alleles. Although all these samples were unmixed based on the PCR-SSOP point mutation genotype data, 99 (14%) isolates contained > 1 allele at one or more microsatellite loci, indicating existence of multiple P. falciparum clones while also highlighting higher sensitivity of microsatellite in detecting polyclonality compared to PCR-SSOP method. A further 183 (25%) isolates did not amplify at one or more microsatellite loci making construction of haplotype incomplete. Both the polyclonal, and isolates with incomplete microsatellite data were excluded from haplotype construction and any further analysis. Overall, 446 (61%) isolates, were included in haplotype construction and analysis of diversity values; 78 were the dhfr sensitive, 184 double mutant dhfr (C59R + S108N or N51I + S108N) and 184 the triple mutant dhfr (N51I + C59R + S108N).
Number of alleles (A) and allele length range at the three microsatellite loci in Kilombero-Ulanga and Rufiji populations
Size range (bp)
Size range (bp)
Size range (bp)
Three-locus microsatellite haplotypes were constructed from the 446 unmixed isolates, in the sensitive, double mutant (C59R+S108N), double mutant (N51I+S108N) and triple mutant dhfr groups (see Additional file 1). A total of 104 different haplotypes were identified, ranked and numbered according to size of allele at -0.3 Kb then at -4.4 Kb and finally at -5.3 Kb locus and numbered from H1-H104. Of the 104 different microsatellite haplotypes, 56 were identified among the 78 sensitive isolates, 19 among the 87 C59R+S108N double mutant isolates, 26 among the 97 N51I+S108N double mutant isolates and 21 were identified among the 184 triple mutant dhfr isolates. Nine haplotypes (H16, H22, H32, H35, H36, H39, H47, H51 and H53) were shared differently between sensitive and double mutant and triple mutant dhfr allele (Additional file 1). Two others (H45 and H52) found among 19 C59R+S108N double mutant isolates were shared with the N51I+S108N double mutant dhfr. H90 was in the majority of the triple mutant dhfr isolates (135 of 184) but was also found in 2 of 106 among N51I+S108N double mutant dhfr isolates. H89 was shared (2 of 106 against 2 of 184) among the N51I+S108N double mutant and the triple mutant dhfr isolates, respectively. This large extent of haplotype sharing is probably a reflection of high degree of recombination in this area of stable high malaria transmission and is consistent with a similar finding in Kenya . The findings reported here confirms recent findings from Kilifi Kenya , reporting high levels of microsatelite haplotype recombination.
Beside the predominant haplotype H90, the triple mutant dhfr had 21 other microsatellite haplotypes, and these differed at one or two of the three flanking microsatellite loci. These alternative haplotypes were generally of low frequency of < 2% (except haplotype 21 which occurred at 11% frequency) and most likely to be relatives of the dominant haplotype with the different microsatellite allele(s) possibly introduced by recombination with the double dhfr mutant types. An exception however, is the existence of four haplotypes (H32, H35, H18 and H38), which had the 87 bp allele at the closest dhfr microsatellite loci (-0.3 Kb). The 87 bp allele at -0.3 Kb loci is a common allele for the sensitives and double dhfr mutants, while 108 bp is for the triple dhfr mutants, thus there is a possibility that these haplotypes represent cases where double mutants have acquired an extra mutation transforming to triple dhfr mutant types.
Two novel haplotypes (H29 and H58) associated with the triple mutant type were identified and differed by allele length with the dominant (Asian type) haplotype at all three microsatellite loci. They have not been described before and in this study each was present in one isolate only. The haplotypic diversity displayed by the double mutant dhfr isolates was intermediate between that found in the sensitive and triple mutant dhfr types. Both double mutant dhfr alleles were associated with four common haplotypes, the C59R+S108N with H35, H51, H34 and H64 and N51I+S108N with H22, H32, H39 and H51. H64 was the most common for the C59R+S108N double mutant, while H51 was the most common for the N51I+S108N double mutant dhfr. Both haplotypes were reported in Tanzania and South Africa  and later found in other African P. falciparum populations [13, 14, 17–19].
The temporal relationship comparing triple mutant dhfr allele heterozygosity values between Kilombero-Ulanga and Rufiji populations generated interesting findings. In Rufiji, He was 0.34 in 2001/02, but increased slightly to 0.55 in the 2006 while in Kilombero-Ulanga, He was 0.65 in 2001/02 but decreased dramatically to 0.23 in 2006. These findings indicate that while escalation of SP drug pressure in Kilombero-Ulanga led to dramatic loss of gene diversity, SP+AS in Rufiji was associated with halt of the loss and stabilization of gene diversity of the triple mutant dhfr allele. Both the sensitive and double mutant dhfr alleles did not seem to respond to either SP or SP+AS drug pressure.
The study indicate that the escalation of dhfr triple mutant resistance allele between 2001/02 - 2006 in Kilombero-Ulanga and Rufiji populations coincided with the change of Nationla anti-malarial policy from CQ first line to SP in late 2001/02. In fact, this observation is concordant to the results of SP treatment efficacy monitoring in southeast Tanzania during 2003, which found that 49% of SP treatments failed by day 28 . It is worth noting that, the observed escalation of dhfr triple mutant resistance allele in the study districts, was the continuation of already established pyrimethamine resistance, following 18 years of SP use as a second-line treatment since 1983. By the time SP was officially replacing CQ as national first line treatment for uncomplicated malaria, the frequencies of the dhfr triple mutant resistance allele were 31% and 47% in Kilombero-Ulanga and Rufiji districts, respectively. As the SP became widely used between late 2001/02-2006, drug pressure intensified and the triple mutant dhfr (N51I+C59R+S108N) allele escalated by a > 2-fold increase in Kilombero-Ulanga population. In Rufiji population the SP drug pressure that was initiated by the change of National anti-malarial guideline in late 2001/02 was interrupted by introduction of SP+AS combination in late 2002. It is widely believed that ACT acts so rapidly killing parasite biomass irrespective of their resistant properties, potentially reducing the likelihood that resistant parasites spread within the community as well as potentially reducing overall malaria transmission rates [34, 35]. While the combination of SP+AS is not recommended for use in areas where SP cure rates are lower than 80%, the present study indicate that SP+AS combination slowed the escalation of resistant dhf r alleles relative to Kilombero-Ulanga, from 11% higher in Rufiji in 2002 to 1% lower in 2006 relative to Kilombero-Ulanga, even though the SP cure rate was lower than 51% in the population .
While the growth of triple mutant allele in Kilombero-Ulanga can be explained by the expansion of the triple mutant dhfr lineage under selection by SP treatment, the failure of SP+AS combination to halt the increase of triple mutant dhfr allele frequency in Rufiji requires some explanations. There are several key ways in which the application of drug pressure solely through provision of SP+AS through government health facilities could have been undermined. Firstly, in Tanzania the availability of drugs is unregulated and varieties of anti-malarial drugs were widely available in shops and kiosks for self-medication during the time of this study . Secondly, the unmatched half-life of 5-10 days for the SP and 45 minutes for the AS leaves little chance for two component drugs to work synergistically as a combination. Consequently, any new infection acquired after AS is cleared in the blood will likely be exposed to SP monotherapy providing opportunity for SP selection. Thirdly, access to treatment was imperfect, only a proportion of infections were treated as a result of high proportions of untreated asymptomatic infections in this area of high and stable malaria transmission. In the present study, the results suggests on average 20% of healthy looking individuals were infected with P. falciparum malaria without expressing clinical symptoms of the disease in Rufiji (see Table 2). This provides a very significant reservoir of parasite populations that are not exposed to chemotherapy [37, 38]. The effect of the reservoir is that, it provides refugia for parasites which are untreated which compete with those exposed to ACT. Fourthly, by the time ACT begun in Rufiji in late 2002, the dhfr triple mutant allele frequency had risen to approximately 63% compromising the possibility of mutual protection of the component drugs in the ACT, the SP and AS. Lastly, While National treatment policy change was found to have a significant major effect on the frequency of resistance alleles, the switch to ACT was applied in one district only, and genetic exchange with surrounding areas which were still subject to selection by SP monotherapy may have contributed to the dilution of the effects of the intervention. This suggestion is supported by studies using dhfr flanking micosatellite  indicating that once established resistant dhfr alleles are highly mobile and this observation was confirmed using similar microsatellite markers in the present study.
Alleles with intermediate resistance, double dhfr mutant N51I+S108N and C59R+S108N did not seem to respond to the strong selection pressure exerted by intensified SP use in Kilombero-Ulanga nor to the apparently reduced SP pressure by the use of ACT in Rufiji, presenting with a constant frequency of 10% throughout in both populations. Identification of mutation at codon 164 of dhfr is a matter of great concern. This novel resistant allele occurred as a minority mixture in majority sensitive dhfr allele and in only a single isolate in this study, suggesting that this highly resistant dhfr genotype allele is beginning to appear but presently, it is still very rare. In the previous study conducted on the same area, around 7,000 P. falciparum-positive isolates from both symptomatic and asymptomatic patients were genotyped but none contained the mutant I164L dhfr mutation. These findings corroborate recent similar findings in different malaria endemic areas of east Africa, Muheza-Tanzania , Uganda [11, 14], Malawi , and Kenya , which reported the I164L mutation at a low frequency, and warn of increasing appearance of this highly resistant P. falciparum in Africa. Most encouraging perhaps is the fact that, currently most of East African countries have already replaced SP with artemisinin-based combination therapies (ACT) as first-line anti-malarial treatment in their respective national guidelines. This will markedly reduce SP usage, which is the key determinant of its selection strength hence limiting either further selection of I164L mutation or establishment of its dispersal mechanism in the population.
Assessment of the evolutionary origin of the resistant dhfr alleles associated with pyrimethamine resistance in P. falciparum isolates from an area of high malaria transmission in two districts of south eastern part of Tanzania has confirmed some previous findings, but also generated new observations which are important for understanding the evolution of resistance at dhfr in African populations. The study detected high genetic diversity in the sensitive dhfr allele supporting the view that this is the ancestral form of dhfr [15, 16]. This finding is consistent with previous studies elsewhere [13, 14, 16–20, 30]. The predominant triple mutant dhfr flanking microsatellite haplotype observed in this study (H90: 108 bp/176 bp/203 bp) was identical to the Asian type haplotype reported in  and subsequent studies indicated it is widespread in Africa [13, 14, 17–20, 39, 40].
The findings of this study shows that, 25% of triple mutant dhfr flanking microsatellite haplotypes shared similar allele length with haplotype H90: 108 bp/176 bp/203 bp at one or two microsatellite loci, indicating that the variation observed are possibly explained by the evolution of H90: 108 bp/176 bp/203 bp through replication errors (mutation)  or recombination as reported in  rather than because of a local independent emergence. Two haplotypes (H29: 87 bp/174 bp/207 bp and H58: 87 bp/184 bp/199 bp), had completely different allele length at all three microsatellite loci and exhibited variation by 1 or 10 repeats compared to haplotype H90: 108 bp/176 bp/203 bp, and in this case it suggests they have arisen independently. Both haplotypes were carrying an allele which is characteristically associated with the dhfr double mutant at the closest microsatellite (-0.3 Kb) loci, and this indicates that these haplotypes may have arisen through recombination between dhfr double mutant and dhfr triple mutant flanking sequences after the unique dhfr haplotype H90: 108 bp/176 bp/203 bp had undergone series of replication errors at the -4.4 Kb and -5.3 Kb loci leading to the 1 or 2 repeat variations observed at the two distant microsatellite loci.
Confirmation of this interpretation may require genotyping of these isolates at other polymorphic sequences that are closely linked to the dhfr gene. Assessment of the dhfr double mutant flanking microsatellite sequences revealed intermediate levels of diversity with sequence variation being more conserved than the sensitives but more diverse than the triple mutant. The two dhfr double mutant alleles (C59R+S108N and N51I+S108N), each had four different common haplotypes indicating that each has arisen independently at least four times, consistent with previous reports [15–17]. A caution however, has to be taken when interpreting haplotype H51: (87 bp/182 bp/193 bp) which was common to both C59R+S108N and N51I+S108N dhfr double mutant alleles. This haplotype was the most common for the N51I+S108N dhfr double mutant allele in the present study and is consistent with earlier findings in the Tanzanian parasite population . It could be a result of mutation, recombination or both, involving the two distant loci (-4.4 Kb and -5.3 Kb) of the C59R+S108N dhfr double mutant flanking microsatellite sequences. Generally, there was a high degree of sharing of both allele length at different microsatellite loci among the dhfr double and triple mutant alleles, highlighting high levels of recombination and errors of replication associated with extreme malaria transmission pattern of the study area. The high level of recombination detected in our study is consistent with recent finding in Kilifi Kenya , likely reflecting high malaria transmission in the eastern African zone.
Taken together, data from this study show that all 184 dhfr triple mutant isolates typed (perhaps with 2 possible exceptions) are related to the Asian ancestral type previously described in  and later found to be dispersed widely in Africa [13, 14, 39, 40]. A study in Western Kenya has also found additional dhfr triple mutant novel haplotypes suggesting multiple origin of the dhfr triple mutant allele . Further studies may be necessary to explore the possibility of replication of Kenyan findings. The findings of the present study however, is very similar to reports of recent studies in Uganda , Kenya  and that of Maiga and co-workers involving samples from 11 sub-Saharan African countries with different levels of malaria transmission . All these studies identified a common evolutionary history of the dhfr triple mutant allele most likely indicating that this unique lineage is responsible for antifolate resistance throughout African region.
Theory suggests SP drug pressure is the driving force behind evolution of its resistance. Positive selective pressure act to increase the frequency of favoured allele meanwhile creating association with the sequences immediately flanking the gene. Initially there is generalized association but as frequency of the favoured allele increases recombination breaks down the more distant associations retaining only associations of the allele with sequences immediately flanking it, the hitchhiker [41, 42]. Ultimately, the signature of selection is a pattern of reduced gene diversity or expected heterozygosity, He and the loss of diversity is described as selective sweep.
In the present study, the progression of triple mutant dhfr allele frequency decreased from 11% higher in Rufiji in 2002 to 1% lower in 2006 relative to Kilombero-Ulanga, clearly indicating reduction of SP drug pressure by the ACT application in Rufiji. Comparison of heterozygosity (He) values for the triple mutant dhfr allele (C59R+N51I+S108N) after four years of SP+AS implementation in Rufiji and four years of SP monotherapy in Kilombero-Ulanga revealed a clear signature of SP selection in Kilombero-Ulanga and a recovery of loss of gene diversity in Rufiji. The diversity values dropped from 0.65 to 0.23 in Kilombero-Ulanga population between 2001/02 and 2006 while in Rufiji the diversity around the allele recovered from 0.34 to 0.55 between 2001/02 and 2006 indicating that, the ACT attributable reduction of SP selection pressure allowed the recovery of gene diversity values through recombination (Figures 5 and 6). The observed significant reduction of gene diversity in Kilombero-Ulanga where SP selection was taking place since 2001 is consistent with previous reports in parasites from south east Africa showing significant loss of diversity across a region of 70 kb around the triple mutant dhfr allele as an evidence of a selective sweep attributable to selection through widespread use of SP for the treatment of malaria .
Finally, like most of preceding studies, the present study has also shown that pyrimethamine resistance is driven by the expansion of the unique Asian triple mutant dhfr allele lineage and recombination and errors of replication were largely responsible for the erosion of this lineage. This, therefore, emphasizes the role of gene flow in the dispersal of parasite population as suggested earlier [15, 17] calling for continental policies to control the drug resistant malaria.
The authors are grateful to all people who participated in the prevalence surveys in Kilombero, Ulanga and Rufiji districts. The Interdisciplinary Monitoring Project for Antimalarial Combination Therapy in Tanzania (IMPACT-Tz) is funded by USAID, CDC, and Wellcome Trust. The co-Principal Investigators are Salim Abdulla and Peter Bloland. Cally Roper and Richard Pearce are supported by a Wellcome Trust Fellowship (ref 060714) awarded to CR. The authors also thank the staff who carried out the cross sectional surveys and the support staff of IHI. The sponsors of this study had no role in study design, data collection, data analysis, data interpretation, or writing of the report.
- Bzik DJ, Li WB, Horii T, Inselburg J: Molecular cloning and sequence analysis of the Plasmodium falciparum dihydrofolate reductase-thymidylate synthase gene. Proc Natl Acad Sci USA. 1987, 84: 8360-8364. 10.1073/pnas.84.23.8360.PubMed CentralView ArticlePubMedGoogle Scholar
- Triglia T, Cowman AF: Primary structure and expression of the dihydropteroate synthetase gene of Plasmodium falciparum. Proc Natl Acad Sci USA. 1994, 91: 7149-7153. 10.1073/pnas.91.15.7149.PubMed CentralView ArticlePubMedGoogle Scholar
- Brooks DR, Wang P, Read M, Watkins WM, Sims PF, Hyde JE: Sequence variation of the hydroxymethyldihydropterin pyrophosphokinase: dihydropteroate synthase gene in lines of the human malaria parasite, Plasmodium falciparum, with differing resistance to sulfadoxine. Eur J Biochem. 1994, 224: 397-405. 10.1111/j.1432-1033.1994.00397.x.View ArticlePubMedGoogle Scholar
- Zolg JW, Plitt JR, Chen GX, Palmer S: Point mutations in the dihydrofolate reductase-thymidylate synthase gene as the molecular basis for pyrimethamine resistance in Plasmodium falciparum. Mol Biochem Parasitol. 1989, 36: 253-262. 10.1016/0166-6851(89)90173-4.View ArticlePubMedGoogle Scholar
- Wu Y, Kirkman LA, Wellems TE: Transformation of Plasmodium falciparum malaria parasites by homologous integration of plasmids that confer resistance to pyrimethamine. Proc Natl Acad Sci USA. 1996, 93: 1130-1134. 10.1073/pnas.93.3.1130.PubMed CentralView ArticlePubMedGoogle Scholar
- Trigg JK, Mbwana H, Chambo O, Hills E, Watkins W, Curtis CF: Resistance to pyrimethamine/sulfadoxine in Plasmodium falciparum in 12 villages in north east Tanzania and a test of chlorproguanil/dapsone. Acta Trop. 1997, 63: 185-189. 10.1016/S0001-706X(96)00617-1.View ArticlePubMedGoogle Scholar
- Wang P, Lee C-S, Bayoumi R, Djimde A, Doumbo O, Swedberg G, Dao Le D, Mshinda H, Tanner M, Watkins W, Sims PFG, Hyde JE: Resistance to antifolates in Plasmodium falciparum monitored by sequence analysis of dihydropteroate synthetase and dihydrofolate reductase alleles in a large number of field samples of diverse origins. Mol Biochem Parasitol. 1997, 89: 161-177. 10.1016/S0166-6851(97)00114-X.View ArticlePubMedGoogle Scholar
- Triglia T, Wang P, Sims PF, Hyde JE, Cowman AF: Allelic exchange at the endogenous genomic locus in Plasmodium falciparum proves the role of dihydropteroate synthase in sulfadoxine-resistant malaria. EMBO (Eur Mol Biol Organ) J. 1998, 17: 3807-3815.View ArticleGoogle Scholar
- Hyde JE: The dihydrofolate reductase-thymidylate synthetase gene in the drug resistance of malaria parasites. Pharmacol Ther. 1990, 48: 45-59. 10.1016/0163-7258(90)90017-V.View ArticlePubMedGoogle Scholar
- Hastings IM, Watkins WM, White NJ: The evolution of drug-resistant malaria: the role of drug elimination half-life. Philos Trans R Soc London Ser B. 2002, 357: 505-519. 10.1098/rstb.2001.1036.View ArticleGoogle Scholar
- Steadke SG, Sendagire H, Lamola S, Kamya MR, Dorsey G, Rosenthal PJ: Relationship between age, molecular markers, and response to sulphadoxine-pyrimethamine treatment in Kampala, Uganda. Trop Med Int Health. 2004, 9: 624-629. 10.1111/j.1365-3156.2004.01239.x.View ArticleGoogle Scholar
- Alker AP, Mwampasa V, Purfield A, Rogerson SJ, Molyneux ME, Kamwendo DD, Tadesse E, Chaluluka E, Meshnick SR: Mutations associated with sulfadoxine-pyrimethamine and chlorproguanil resistance in Plasmodium falciparum isolates from Blantyre, Malawi. Antimicrob Agents Chemother. 2005, 49: 3919-3921. 10.1128/AAC.49.9.3919-3921.2005.PubMed CentralView ArticlePubMedGoogle Scholar
- McCollum AM, Poe AC, Hamel M, Huber C, Zhou Z, Shi YP, Ouma P, Vulule J, Bloland P, Slutsker L, Barnwell JW, Udhayakumar V, Escalante AA: Antifolate resistance in Plasmodium falciparum: multiple origins and identification of novel dhfr alleles. J Infect Dis. 2006, 194: 189-197. 10.1086/504687.View ArticlePubMedGoogle Scholar
- Lynch C, Pearce R, Pota H, Cox J, Abeku TA, Rwakimari J, Naidoo I, Tibenderana J, Roper C: Emergence of a dhfr mutation conferring high-level drug resistance in Plasmodium falciparum populations from southwest Uganda. J Infect Dis. 2008, 197: 1598-1604. 10.1086/587845.View ArticlePubMedGoogle Scholar
- Roper C, Pearce R, Bredenkamp B, Gumede J, Drakeley C, Mosha F, Chandramohan D, Sharp B: Antifolate antimalarial resistance in southeast Africa. Lancet. 2003, 361: 1174-81. 10.1016/S0140-6736(03)12951-0.View ArticlePubMedGoogle Scholar
- Roper C, Pearce R, Nair S, Sharp B, Nosten F, Anderson T: Intercontinental spread of pyrimethamine resistant malaria. Science. 2004, 305: 1124-10.1126/science.1098876.View ArticlePubMedGoogle Scholar
- Pearce R, Malisa A, Kachur SP, Barnes K, Sharp B, Roper C: Reduced Variation Around Drug Resistant dhfr Alleles in African Plasmodium falciparum. Mol Biol Evol. 2005, 22: 1834-1844. 10.1093/molbev/msi177.View ArticlePubMedGoogle Scholar
- Ndiaye D, Daily JP, Sarr O, Ndir O, Gaye O, Mboup S, Roper C, Wirth DF: Defining the origin of Plasmodium falciparum resistant dhr isolates in Senegal. Acta Trop. 2006, 99: 106-111. 10.1016/j.actatropica.2006.07.002.PubMed CentralView ArticlePubMedGoogle Scholar
- Maiga O, Djimde AA, Hubert V, Renard E, Abouy A, Kironde F, Nsimba B, Koram K, Doumbo OK, Le Bras J, Clain J: A shared Asian Origin of the Triple-Mutant dhfr Allele in Plasmodium falciparum from sites across Africa. J Infect Dis. 2007, 196: 165-172. 10.1086/518512.View ArticlePubMedGoogle Scholar
- Certain LK, Briceno M, Kiara SM, Nzila AM, Watkins WM, Sibley CH: Characteristics of Plasmodium falciparum dhfr haplotypes that confer pyrimethamine resistance, Kilifi, Kenya, 1987-2006. J Infect Dis. 2008, 197: 1743-51. 10.1086/588198.PubMed CentralView ArticlePubMedGoogle Scholar
- Hay SI, Rogers DJ, Toomer JF, Snow RW: Annual Plasmodium falciparum entomological inoculation rates (EIR) across Africa: literature survey, Internet access and review. Trans R Soc Trop Med Hyg. 2001, 94: 113-127.View ArticleGoogle Scholar
- Pearce R, Drakeley C, Chandramohan D, Mosha F, Roper C: Molecular determination of point mutation haplotypes in the dihydrofolate reductase and dihydropteroate synthase of Plasmodium falciparum in three districts of northern Tanzania. Antimicrob Agents Chemother. 2003, 47: 1347-1354. 10.1128/AAC.47.4.1347-1354.2003.PubMed CentralView ArticlePubMedGoogle Scholar
- Stata Corp: Stata Statistical Software: Release 7.0. 2002, College Station, TX: Stata CorporationGoogle Scholar
- Mugittu K, Ndejembi M, Malisa A, Lemnge M, Premji Z, Mwita A, Nkya W, Kataraihya J, Abdulla S, Beck H, Mshinda H: Therapeutic efficacy of sulphadoxine-pyrimethamine and prevalence of resistance markers in Tanzania prior to revision of malaria treatment policy: Plasmodium falciparum dihydrofolate reductase and dihydropteroate synthase mutations in monitoring in vivo resistance. Am J Trop Med Hyg. 2004, 71: 696-702.PubMedGoogle Scholar
- Malisa A, Pearce RJ, Abdulla S, Mshinda H, Kachur P, Bloland P, Roper C: Drug coverage in treatment of malaria and the consequences for resistance evolution - evidence from the use of sulphadoxine/pyrimethamine. Malar J. 2010, 9: 190-10.1186/1475-2875-9-190.PubMed CentralView ArticlePubMedGoogle Scholar
- Bwijo B, Kaneko A, Takechi M, Zungu IL, Moriyama Y, Lum JK, Tsukahara T, Mita T, Takahashi N, Bergqvist Y, Bjorkman A, Kobayakawa T: High prevalence of quintuple mutant dhfr/dhps genes in Plasmodium infections seven years after introduction of sufadoxine and pyrimethamine as first line treatment in Malawi. Acta Trop. 2003, 85: 363-373. 10.1016/S0001-706X(02)00264-4.View ArticlePubMedGoogle Scholar
- Nzila AM, Mberu EK, Dayo H, Winstanley PA, Sibley CH, Watkins WM: Towards an understanding of the mechanism of pyrimethamine-sulfadoxine resistance in Plasmodium falciparum: Genotyping of dihydrofolate reductase and dihydropteroate synthase of Kenyan parasites. Antimicrob Agents Chemother. 2000, 44: 991-996. 10.1128/AAC.44.4.991-996.2000.PubMed CentralView ArticlePubMedGoogle Scholar
- Omar SA, Adagu IS, Warhurst DC: Can pretreatment screening for dhps and dhfr point mutations in Plasmodium falciparum infections be used to predict sulfadoxine-pyrimethamine treatment failure?. Trans R Soc Trop Med Hyg. 2001, 95: 315-319. 10.1016/S0035-9203(01)90250-0.View ArticlePubMedGoogle Scholar
- Kyabayinze D, Cattamanchi A, Kamya MR, Rosenthal PJ, Dorsey G: Validation of a simplified method for using molecular markers to predict sulfadoxine-pyrimethamine treatment failure in African children with falciparum malaria. Am J Trop Med Hyg. 2003, 69: 247-252.PubMedGoogle Scholar
- Sendagire H, Kyabayinze D, Swedberg G, Kironde F: Plasmodium falciparum: higher incidence of molecular resistance markers for sulphadoxine than for pyrimethamine in Kasangati, Uganda. Trop Med Int Health. 2005, 10: 537-543. 10.1111/j.1365-3156.2005.01414.x.View ArticlePubMedGoogle Scholar
- Anderson TJ, Haubold B, Williams JT, Estrada-Franco JG, Richardson L, Mollinedo R, Bockarie M, Mokili J, Mharakurwa S, French N, Whitworth J, Velez ID, Brockman AH, Nosten F, Ferreira MU, Day KP: Microsatellite markers reveal a spectrum of population structures in the malaria parasite Plasmodium falciparum. Mol Biol Evol. 2000, 17: 1467-1482.View ArticlePubMedGoogle Scholar
- Anderson TJ: Mapping drug resistance genes in P. falciparum by genome-wide association. Curr Drug Targets Infect Disord. 2004, 4: 65-78. 10.2174/1568005043480943.View ArticlePubMedGoogle Scholar
- Mugittu K, Abdulla S, Falk N, Masanja H, Felger I, Mshinda H, Beck H, Genton B: Efficacy of sulphadoxine-pyrimethamine in Tanzania after two years as first-line drug for uncomplicated malaria: assessment protocol and implication for treatment policy. Malar J. 2005, 4: 55-10.1186/1475-2875-4-55.PubMed CentralView ArticlePubMedGoogle Scholar
- Price RN, Nosten F, Luxemburger C: Effects of artemisinin derivatives on malaria transmissibility. Lancet. 1996, 347: 1654-1658. 10.1016/S0140-6736(96)91488-9.View ArticlePubMedGoogle Scholar
- White NJ: Preventing antimalarial drug resistance through combinations. Drug Resist Updat. 1998, 1: 3-9. 10.1016/S1368-7646(98)80208-2.View ArticlePubMedGoogle Scholar
- Goodman C, Kachur SP, Abdulla S, Mwageni E, Nyoni J, Schellenberg JA, Mills A, Bloland P: Retail supply of malaria-related drugs in rural Tanzania: risks and opportunities. Trop Med Int Health. 2004, 9: 655-663. 10.1111/j.1365-3156.2004.01245.x.View ArticlePubMedGoogle Scholar
- Smith G: Chemotherapy: Future problems. Hookworm disease: Current Status and New Directions. Edited by: Schad, GA, Warren KS. 1990, Taylor and Francis, London, 291-303.Google Scholar
- Anderson RM, May RM: Infectious Diseases of Human: Dynamics and Control. 1991, Oxford University press, OxfordGoogle Scholar
- McCollum AM, Basco LK, Tahar R, Udhayakumar V, Escalantes A: Hitchhiking and selective sweeps of Plasmodium falciparum sulfadoxine and pyrimethamine resistance alleles in a population from Central Africa. Antimicrob Agents Chemother. 2000, 52: 4089-4097.View ArticleGoogle Scholar
- Mita T, Tanabe K, Takahashi N, Culleton R, Ndounga M, Dzodzomenyo M, Akhwale WS, Kaneko A, Kobayakawa T: Indigenous evolution of Plasmodium falciparum pyrimethamine resistance multiple times in Africa. Antimicrob Agents Chemother. 2009, 63: 252-255.View ArticleGoogle Scholar
- Smith JM, Haigh J: The hitch-hiking effect of a favourable gene. Genet Res. 1974, 23: 23-35. 10.1017/S0016672300014634.View ArticlePubMedGoogle Scholar
- Kaplan NL, Hudson RR, Langley CH: The "hitchhiking effect" revisited. Genetics. 1989, 123: 887-899.PubMed CentralPubMedGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.