Abstract
We previously reported a high baseline prevalence of mutations in the dhfr and dhps genes of Plasmodium falciparum throughout Senegal. The highest prevalence of the triple dhfr pyrimethamine associated mutations were found in isolates obtained in the western part of the country near the capital city of Dakar. In this study, we sought out to determine the relatedness of dhfr wild type and mutated strains by analyzing three microsatellite regions upstream of the dhfr locus. Twenty-six of the 31 wild type strains had a unique microsatellite pattern. In contrast of the 17 isolates containing the triple mutation in dhfr, 11 had an identical microsatellite pattern. Diverse geographical isolates in Senegal containing the triple dhfr mutation have arisen from a limited number of ancestral strains. In addition, we demonstrate that these isolates have shared ancestry with the previously reported triple mutation haplotype found in Tanzania, South Africa, southeast Asia. This common ancestry may have implications for the malaria control strategy for reducing the spread of sulfadoxine–pyrimethamine resistance in Senegal and elsewhere in Africa.
Keywords: Plasmodium falciparum, Drug resistance, Senegal, Dhfr, Sulfadoxine–pyrimethamine
1. Introduction
Drug therapy remains the mainstay of management in Plasmodium sp. infections; however, only a limited number of compounds are available and drug resistance is common. Because of high prevalence of chloroquine resistance in Senegal, the national malaria treatment policy was recently changed to sulfadoxine–pyrimethamine-amodiaquine (SP-AQ) from chloroquine. A high prevalence of isolates containing the dhfr pyrimethamine drug resistance haplotype (N51I, C59R, S108N) was described prior to the introduction of SP-AQ as first line therapy in Senegal (Ndiaye et al., 2005). Furthermore, there is a marked difference in the prevalence of resistance across the country. The highest prevalence of dhfr genotypic resistance occurs in the western coastal region near the capital city of Dakar, whereas, a lower prevalence of resistance is found in the eastern, interior region of the country. Upstream flanking microsatellites of dhfr in wild type and genotypic resistant dhfr isolates was analyzed and compared between these two regions to understand the origin and spread of pyrimethamine resistance. Microsatellite size polymorphisms in the upstream flanking region of dhfr are highly variable in wild type isolates as well as in those isolates with one or two mutations in dhfr. In contrast, isolates containing the triple mutation in dhfr are less polymorphic, suggesting they arose from a small number of ancestral strains. In addition, the microsatellite sizes of the strains containing the triple mutation is similar between the two geographic regions and those previously reported from southeast Asia.
2. Material and methods
2.1. Patient population
To define the relationship of parasite isolates at the dhfr locus we analyzed samples from two geographically separated regions in Senegal (Fig. 1). Samples were collected from patients with a blood smear positive for Plasmodium falciparum, mild malaria, 5 years or greater and no history of recent treatment with antimalarials or chronic illness. Forty-one blood samples were analyzed from western Senegal in Pikine and Thies and the eastern villages near Tambacounda and Santhioumaleme which had been collected as part of a previous study in 2003 (Ndiaye et al., 2005). To increase the sample size from eastern Senegal, an additional 18 samples from were collected and analyzed from villages near Velingara in 2004. Patient blood samples were collected after informed consent was obtained. The Human Subjects Committee of Harvard School of Public Health in Boston and the Ethics Committee of Cheikh Anta Diop University in Dakar approved the protocol used in this study.
Fig. 1.
Map of Senegal and study sites. Samples derived from western Senegal were obtained in Pikine and Thies, near the capital city of Dakar. Samples from eastern Senegal were collected from villages near Tambacounda, Santhioumaleme and Velingara.
2.2. Study sites
In the western part of Senegal, the study region included Pikine, which is 15 km from the coastal city of Dakar and has an entomological inoculation rate (EIR) of 1 as well as Thies (EIR = 1–20), which is 70 km from Dakar (Trape et al., 1992). The second study region is greater than 400 km from Dakar in the eastern portion of the country and has, in contrast to the west, a high transmission rate with EIR > 100 in the small villages surrounding Santhioumaleme, Tambacounda and Velingara (Faye et al., 1995).
2.3. Molecular analysis
DNA was extracted from finger prick samples collected on filter paper using DNA Minikit (Quiagen) following the manufacturer's instructions and used for dhfr sequencing and microsatellite analysis. Dhfr haplotypes were already defined for all villages except Velingara. To determine the prevalence of dhfr mutations from the Velingara 2004 samples, nested PCR was followed by cloning and sequencing using primers and protocol as previously described (Ndiaye et al., 2005). To determine the number of clones in each sample, genotyping was performed for the merozoite surface protein (MSP) 1 and MSP 2 alleles by use of polymerase chain reaction (PCR) with nested primers as previously described (Snounou et al., 1999).
2.4. Microsatellite analysis
Microsatellites at 0.3, 4.4 and 5.3 kb upstream of dhfr gene were analyzed based on Roper et al. primer sequences and protocol (Roper et al., 2003). Additional nucleotide sequence were added to three of these primers to minimize stutter artifact: PFDHFR0.3 kb (F: 5′-CTG TCT TAT TCC AAC ATT TTC AAG A-3′) PFDHFR4.4 kb (R: 5′-CTG TCT TCG ATA TAT CTG ATG GGT GA-3′); PFDHFR5.3 kb (R: 5′-CTG TCT TCA CAT ATT ATA CAG GAC G A-3′). PCR amplification resulted in single PCR bands by agarose gel electrophoresis and the nucleotide sequences were confirmed after cloning and sequencing. Microsatellite size was determined by capillary gel electrophoresis of the PCR product using an Applied Biosystems 3730 × l DNA Analyzer. Genemapper version 3.5 (ABI) was used to analyze the data and the highest peak is reported in this study for statistical comparisons. To determine if the triple mutant haplotype from Senegal shared ancestry with the previously reported SE Asian, South African and Tanzanian triple mutation haplotype, the samples were simultaneously analyzed with SE Asian and African standards on an identical ABI Analyzer in Dr. Roper's laboratory. The primer sequences for this analysis were taken directly from Roper et al. (2003) published protocol without modification.
2.5. Statistical analysis
To compare the statistical association of microsatellite size polymorphisms and dhfr drug resistance haplotypes Fisher's exact test was used for two-tailed significance at p = 0.05 (STATA, 9.0, Stata Corporation).
3. Results
The additional haplotype data from villages near Velingara revealed 72% of the samples to contain wild type dhfr. Of the 18 samples collected, 13 were wild type, 1 had a single mutation at S108N, and 4 had the triple mutation (N51I, C59R, S108N).
To evaluate ancestry of the triple dhfr mutation isolates, we analyzed microsatellites upstream of dhfr in 59 samples. Thirty samples were derived from western Senegal and 29 samples came from eastern regions of Senegal. The dhfr haplotypes included 31 wild type, 11 single mutation samples (S108N) or double mutation (C59R–S108N) and 17 samples containing triple mutations (N51I, C59R, S108N).
Over 50% of the samples are monoclonal within each dhfr haplotype by MSP1 and 2 genotyping. There are 34 samples with monoclonal infection and the number of monoclonal samples in each dhfr haplotype is similar. Compared to the number of monoclonal infections in the dhfr wild type samples, the number of monoclonal infections in the one to two mutation group or the triple mutation group was not significantly different (Fisher's exact test p = 1, p = 0.8, respectively).
Table 1a reports the microsatellite sizes for the 31 dhfr wild type samples from western and eastern Senegal. The microsatellite 0.3 kb upstream from start site of dhfr ranges in size from 93 to 118 bps, marker 4.4 kb ranges from 175 to 188 bps and marker 5.3 kb varies between 187 and 225 bps. The 4.4 kb microsatellite demonstrates the least size variability with 75% of the samples containing 183 bp. When analyzing all three microsatellite polymorphisms, there are 26 unique microsatellite patterns within the wild type isolates. The haplotype, 109/183/211 is found once in the western region and twice in the eastern region. Of the 11 samples containing one (S108N) or two dhfr mutations (C59R–S108N), all have unique microsatellite patterns (Table 1b). In the 17 samples that contain the triple dhfr mutation, there are four patterns, with the majority containing the 109/183/211 haplotype (Table 2). The occurrence of the 109/183/211 pattern is not statistically different between wild type and one to two mutations haplotypes. Conversely there is statistically significant higher prevalence of this pattern in the triple mutation isolates compared to the samples containing the wild type haplotype (Fisher's exact test, two-sided p = 0.000). This remains statistically significant when samples that are monoclonal by MSP1/2 only are analyzed (p = 0.011).
Table 1a.
Microsatellite sizes of wild type dhfr isolates in Senegal
| sample | Dhfr0.3kb | Dhfr4.4kb | Dhfr5.3kb |
|---|---|---|---|
| P33.03 | 93 | 183 | 211 |
| Th6.03 | 103 | 183 | 200 |
| P37.03 | 103 | 183 | 211 |
| P44.03 | 106 | 177 | 214 |
| Th4.03 | 109 | 183 | 211 |
| P9.03 | 109 | 183 | 187 |
| Th16.03 | 109 | 183 | 217 |
| P45.03 | 109 | 183 | 221 |
| Th9.03 | 113 | 175 | 211 |
| P49.03 | 113 | 183 | 211 |
| P78.03 | 113 | 183 | 211 |
| P17.03 | 116 | 183 | 204 |
| P20.03 | 118 | 183 | 211 |
| T6.03 | 93 | 183 | 211 |
| S5.03 | 93 | 183 | 217 |
| T7.03 | 93 | 183 | 221 |
| V32.04 | 93 | 186 | 217 |
| V35.04 | 100 | 188 | 211 |
| V28.04 | 109 | 183 | 211 |
| V48.04 | 109 | 183 | 211 |
| T1.03 | 109 | 183 | 214 |
| S1.03 | 109 | 185 | 208 |
| V25.04 | 113 | 183 | 211 |
| V40.04 | 113 | 183 | 211 |
| V51.04 | 113 | 183 | 217 |
| V13.04 | 113 | 183 | 221 |
| V37.04 | 113 | 183 | 225 |
| V55.04 | 116 | 193 | 214 |
| V20.04 | 116 | 193 | 221 |
| V43.04 | 118 | 183 | 208 |
| V50.04 | 118 | 188 | 208 |
Three microsatellites upstream of the start site of dhfr at 0.3, 4.4 and 5.3 kb are reported. Two geographic regions are studied, with non-bolded sample names from western Senegal: P = Pikine, Th = Thies; below the dark line and bolded are isolates collected from eastern Senegal: T = Tambacounda, S = Santhioumaleme, V = Velingara. Microsatellites sizes that dominate the triple mutation are in grey (109, 183, 211). Note unique patterns in most isolates.
Table 1b.
Microsatellite sizes of single or double mutation dhfr isolates in Senegal
| sample | Dhfr0.3kb | Dhfr4.4kb | Dhfr5.3kb |
|---|---|---|---|
| **P21.03 | 93 | 188 | 200 |
| *Th12.03 | 103 | 183 | 200 |
| *P24.03 | 103 | 183 | 218 |
| *P25.03 | 109 | 179 | 218 |
| *Th 10.03 | 109 | 183 | 211 |
| *P22.03 | 118 | 179 | 218 |
| *S18.03 | 93 | 183 | 221 |
| *V34.04 | 103 | 188 | 200 |
| **S7.03 | 103 | 183 | 211 |
| *T4.03 | 109 | 183 | 218 |
| **S13.03 | 109 | 183 | 208 |
Samples denoted with contain one mutation (S108N).
Contains two mutations (C59R–S108N), note unique patterns in all isolates. For more details see the legend of Table 1a.
Table 2.
Microsatellite sizes of triple (N51I, C59R, S108N) mutation dhfr isolates in Senegal
| sample | Dhfr0.3kb | Dhfr4.4kb | Dhfr5.3kb |
|---|---|---|---|
| P19.03 | 93 | 183 | 211 |
| P51.03 | 93 | 183 | 211 |
| P13.03 | 109 | 183 | 211 |
| P14.03 | 109 | 183 | 211 |
| P15.03 | 109 | 183 | 211 |
| P35.03 | 109 | 183 | 211 |
| P42.03 | 109 | 183 | 211 |
| P46.03 | 109 | 183 | 211 |
| P53.03 | 109 | 183 | 211 |
| P69 03 | 113 | 183 | 211 |
| Th11.03 | 113 | 183 | 211 |
| S17.03 | 109 | 183 | 211 |
| V21.04 | 109 | 183 | 211 |
| V52.04 | 109 | 183 | 211 |
| S20.03 | 109 | 183 | 225 |
| V54.04 | 113 | 183 | 211 |
| V57.04 | 113 | 183 | 211 |
Three microsatellites upstream of the start site of dhfr at 0.3, 4.4 and 5.3 kb are reported. Two geographic regions are studied, with non-bolded names from western Senegal: P = Pikine, Th = Thies; below the dark line and bolded are isolates collected from eastern Senegal: T = Tambacounda, S = Santhioumaleme, V = Velingara. Microsatellite sizes that dominate the triple mutation are in grey (109, 183, 211). Note the loss of variation in microsatellites sizes, with one pattern, 109/183/211 dominating in both regions.
Samples P14.03 and S17.03 which contain the triple mutation haplotype and the common 109/183/211 microsatellite haplotype were run with the previously reported SE Asian and African triple mutation haplotype samples and found to have matching microsatellite sizes.
4. Discussion
P. falciparum has developed resistance to many classes of antimalarials. The analysis of drug resistance development and its spread is important until non-chemotherapeutic options become available for prevention/treatment of malaria. Our previous studies demonstrated a high prevalence of isolates containing a triple dhfr drug resistance associated haplotype in Senegal. This triple mutation has been shown to correlate with significant antifolate drug resistance (Sibley et al., 2001). As the distance from the capital city, Dakar increases, the prevalence of the dhfr triple mutation decreases (Ndiaye et al., 2005). We wanted to determine the genetic relationship of these triple mutation parasites, as previous reports have found that dhfr drug resistant isolates have arisen from a few primary isolates, rather than occurring from independent mutational events (Nair et al., 2003; Roper et al., 2003, 2004; Pearce et al., 2005; Cortese et al., 2002).
We previously reported that the drug resistance associated dhfr triple mutation haplotype (N551R, C59R, S108N) in western Senegal including Pikine and Thies was 69% (Ndiaye et al., 2005). This was in contrast to a higher prevalence of wild type isolates in eastern Senegal. The additional dhfr haplotype data analyzed from Velingara in the eastern region of Senegal confirms the lower prevalence of dhfr resistance with only 28% of isolates containing the triple dfhr mutation. For the microsatellite analysis, we predicted that the flanking region of dhfr in wild type samples would be diverse, reflecting unique origins, and a lack of linkage disequilibrium. Indeed, the microsatellites at 0.3 and 5.3 kb upstream of dhfr show marked size polymorphism. Interestingly the microsatellite 4.4 kb upstream shows very little size variability and therefore is less informative regarding shared ancestry. Microsatellites can have unique mutation rates based on the length of the dinucleotide repeats. Microsatellites containing long dinucleotide repeats have a higher heterogeneity compared to microsatellite containing shorter repeats (Anderson et al., 2000). The 4.4 kb marker contains the shortest dinucleotide repeat and demonstrates the least heterogeneity compared to the microsatellites at 0.3 and 5.3 kb, which is consistent with this observation. The size variation in the 0.3 and 5.3 kb microsatellites in the wild type isolates suggests that this allele is ancestral. The microsatellites from isolates containing one or two dhfr mutations are also diverse and appear to have arisen independently. The most striking finding is the lack of diversity in the microsatellites of the isolates containing the triple dhfr mutation. There is dominant pattern with 65% of the isolates containing a 109/183/211 pattern and this is identical in both geographic regions. The second most common pattern found in 24% of the triple mutation samples varies only at the 0.3 kb marker (113/183/211), and is also found in both regions.
The lack of diversity in the triple mutant is consistent with the model that parasite mutations become established in the population infrequently, despite their enhanced fitness under drug pressure (Hastings, 2004). Strains with the triple mutation that are highly prevalent may contain additional mutations outside the dhfr locus to allow them to maintain fitness and be transmitted (Walliker et al., 2005). Further genetic analysis of strains with the dhfr triple mutation in high prevalence parasites versus low prevalence parasites may provide insight into mechanisms of parasite fitness and transmission.
The higher prevalence of the triple dhfr mutation in western Senegal suggests that the resistant strains originated in this region before spreading to eastern Senegal. The exact local origin of this drug resistant haplotype in Senegal is unknown, however we have demonstrated shared ancestry of the common microsatellite haplotype 109/183/211 with the triple mutation haplotype found in Tanzania, South Africa, southeast Asia (Roper, 2003, 2004) and most recently Kenya (McCollum et al., 2006). The slightly larger allele sizes reported here are due to differences in primer length used in our protocol, but the match was confirmed by direct comparison. We also observed the triple mutant in association with three less common but related microsatellite backgrounds 93/183/211, 113/183/211 and 109/183/255. Similarly the Kenyan study found two related microsatellite haplotypes associated with the triple mutant also differing at the 0.3 kb locus, together with one entirely different microsatellite haplotype which was unrelated to the ‘Asian’ type (McCollum et al., 2006). Variation in drug pressure, transmission intensity and parasite population diversity may underlie the development of novel drug resistant strains and/or favor spread of alleles with shared ancestry. Further mapping of drug resistant strains in Africa will provide insight into mechanisms of spread and development of drug resistance.
Acknowledgments
A. Gunasekera and D. Milner for careful reading of the manuscript. The work was partially supported by National Institute of Allergy and Infectious Diseases (5K23AI054518 to JPD) and Fogarty International Training grant (5D43TW001503 to DFW).
Footnotes
Uncited reference: Nash et al. (2005).
References
- Anderson TJ, Su XZ, Roddam A, Day KP. Complex mutations in a high proportion of microsatellite loci from the protozoan parasite Plasmodium falciparum. Mol Ecol. 2000;9:1599–1608. doi: 10.1046/j.1365-294x.2000.01057.x. [DOI] [PubMed] [Google Scholar]
- Cortese JF, Caraballo A, Contreras CE, Plowe CV. Origin and dissemination of Plasmodium falciparum drug-resistance mutations in South America. J Infect Dis. 2002;186:999–1006. doi: 10.1086/342946. [DOI] [PubMed] [Google Scholar]
- Faye O, Gaye O, Fontenille D, Sy N, Konate L, Hebrard G, Herve JP, Trouillet J, Diallo S, Mouchet J, et al. Comparison of the transmission of malaria in two epidemiological patterns in Senegal: the Sahel border and the Sudan-type savanna. Dakar Med. 1995;40:201–207. [PubMed] [Google Scholar]
- Hastings IM. The origins of antimalarial drug resistance. Trends Parasitol. 2004;20:512–518. doi: 10.1016/j.pt.2004.08.006. [DOI] [PubMed] [Google 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. doi: 10.1086/504687. [DOI] [PubMed] [Google Scholar]
- Nair S, Williams JT, Brockman A, Paiphun L, Mayxay M, Newton PN, Guthmann JP, Smithuis FM, Hien TT, White NJ, Nosten F, Anderson TJ. A selective sweep driven by pyrimethamine treatment in southeast asian malaria parasites. Mol Biol Evol. 2003;20:1526–1536. doi: 10.1093/molbev/msg162. [DOI] [PubMed] [Google Scholar]
- Nash D, Nair S, Mayxay M, Newton PN, Guthmann JP, Nosten F, Anderson TJ. Selection strength and hitchhiking around two anti-malarial resistance genes. Proc Biol Sci. 2005;272:1153–1161. doi: 10.1098/rspb.2004.3026. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Ndiaye D, Daily JP, Sarr O, Ndir O, Gaye O, Mboup S, Wirth DF. Mutations in Plasmodium falciparum dihydrofolate reductase and dihydropteroate synthase genes in Senegal. Trop Med Int Health. 2005;10:1176–1179. doi: 10.1111/j.1365-3156.2005.01506.x. [DOI] [PMC free article] [PubMed] [Google 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. doi: 10.1093/molbev/msi177. [DOI] [PubMed] [Google Scholar]
- Roper C, Pearce R, Bredenkamp B, Gumede J, Drakeley C, Mosha F, Chandramohan D, Sharp B. Antifolate anti-malarial resistance in southeast Africa: a population-based analysis. Lancet. 2003;361:1174–1181. doi: 10.1016/S0140-6736(03)12951-0. [DOI] [PubMed] [Google Scholar]
- Roper C, Pearce R, Nair S, Sharp B, Nosten F, Anderson T. Intercontinental spread of pyrimethamine-resistant malaria. Science. 2004;305:1124. doi: 10.1126/science.1098876. [DOI] [PubMed] [Google Scholar]
- Sibley CH, Hyde JE, Sims PF, Plowe CV, Kublin JG, Mberu EK, Cowman AF, Winstanley PA, Watkins WM, Nzila AM. Pyrimethamine–sulfadoxine resistance in Plasmodium falciparum: what next? Trends Parasitol. 2001;17:582–588. doi: 10.1016/s1471-4922(01)02085-2. [DOI] [PubMed] [Google Scholar]
- Snounou G, Zhu X, Siripoon N, Jarra W, Thaithong S, Brown KN, Viriyakosol S. Biased distribution of msp1 and msp2 allelic variants in Plasmodium falciparum populations in Thailand. Trans R Soc Trop Med Hyg. 1999;93:369–374. doi: 10.1016/s0035-9203(99)90120-7. [DOI] [PubMed] [Google Scholar]
- Trape JF, Lefebvre-Zante E, Legros F, Ndiaye G, Bouganali H, Druilhe P, Salem G. Vector density gradients and the epidemiology of urban malaria in Dakar, Senegal. Am J Trop Med Hyg. 1992;47:181–189. doi: 10.4269/ajtmh.1992.47.181. [DOI] [PubMed] [Google Scholar]
- Walliker D, Hunt P, Babiker H. Fitness of drug-resistant malaria parasites. Acta Trop. 2005;94:251–259. doi: 10.1016/j.actatropica.2005.04.005. [DOI] [PubMed] [Google Scholar]