Short communication: Point mutations in the dihydrofolate reductase and dihydropteroate synthase genes of Plasmodium falciparum isolates from Colombia


correspondence Dr Tomas Jelinek, Department of Infectious Diseases and Tropical Medicine, University of Munich, Leopoldstr. 5, 80802 Munich, Germany. Fax: +49 89 336 112; E-mail:


Point mutations in the dihydrofolate reductase (DHFR) and dihydropteroate synthase (DHPS) genes of Plasmodium falciparum can lead to an increasing resistance of P. falciparum to pyrimethamine/sulphadoxine. We examined the prevalence of these mutations in 36 samples from Colombia. Analysed by polymerase chain reaction (PCR) for infection with P. falciparum, 25 (69%) tested positive. These positive isolates were tested further for point mutations in the genes of DHFR (codons 16, 51, 59, 108 and 164) and DHPS (codons 436, 437, 540, 581 and 613) by nested PCR and following mutation-specific restriction enzyme digestion. Gene mutations occurred in both the DHFR and DHPS gene of the Colombian isolates, suggesting that resistance to antifolate drugs exists or may develop soon in Colombia.


Malaria is a major public health problem and is associated with high mortality and morbidity rates. The number of cases in Colombia has increased steadily over the past 20 years. Recent studies have shown a high rate of treatment failure with chloroquine in the Pacific coast region of Colombia. A 50% failure of chloroquine was detected in Buenaventura, a port city situated in the central Pacific coast region and a failure rate of >70% was found in El Bagre, Antioquia, in the Magdalena river valley (Malaria Foundation International 2002). Chloroquine-resistant malaria is treated with the combination of the antifolates sulphadoxine and pyrimethamine (S/P), commercially known as Fansidar, as a second line drug. With its widespread use, resistance against this antifolate combination has been developing in different parts of the world (Kondrachine & Trigg 1997). Conditions that encourage the development of resistance are widespread throughout Latin America: over-the-counter sale of antimalarials, frequent self-medication, minimal regulation of chemotherapeutic usage within or outside hospitals, scarce documentation of clinical trial results for newer antimalarials, almost non-existent surveillance and reporting of resistance patterns.

Pyrimethamine targets the enzyme dihydrofolate reductase (DHFR) which catalyses the nicotinamide adenine dinucleotide phosphate-dependent reduction of 7,8-dihydrofolate to 5,6,7,8-tetrahydrofolate (Blakeley 1984). Similarly, dihydropteroate synthase (DHPS), catalysing the formation of dihydropteroate from ρ-aminobenzoic acid (PABA) and 6-hydroxymethyldihydropterine pyrophosphate in the folate biosynthetic pathway (Walter 1991), is the target of sulphadoxine. Evidence exists that point mutations in the DHFR and DHPS genes can be related to resistance to S/P (Brooks et al. 1994; Wang et al. 1997; Jelinek et al. 1998; Plowe et al. 1998). A mutation from serine to asparagine at codon 108 in the DHFR gene is known to cause resistance to pyrimethamine and a moderate loss of response to cycloguanil (the active form of proguanil). Higher levels of pyrimethamine resistance result from additional mutations from asparagine to isoleucine at position 51 and/or from cysteine to arginine at position 59. Polymorphisms in the codons 436, 437, 540, 581 and 613 within the DHPS gene have been reported to cause sulphadoxine resistance. In order to elucidate most of the known alleles of DHFR and DHPS, mutation-specific polymerase chain reaction (PCR) tests have been used (Zolg et al. 1991; Gyang et al. 1992; Wang et al. 1995; Cortese & Plowe 1998; Jelinek et al. 1998). Because of their potential to predict in vivo resistance and therapeutic failure, these molecular methods may become important tools for the surveillance and control of malaria treatment and therefore be useful in public health. Such data can provide crucial information on the potential for S/P resistance in a given area even before the drug combination has been introduced.

Patients, materials and methods


Dried bloodspots were collected from patients with falciparum malaria during a population-based survey in Apartadó (Urabá region, Northern Colombia). Inclusion criteria for the enrolment of patients were the following: older than 18 years; having a history, symptoms or signs compatible with uncomplicated malaria; plasmodial infection, willingness to be followed up. Patients with signs of severe malaria or danger signs (inability to stand or drink, lethargy or unconsciousness, repeated convulsions and persistent vomiting), febrile diseases other than malaria or pregnancy were excluded. Each patient signed an informed consent form. Parasite density was examined microscopically using Giemsa-stained thick and thin smears. Parasite density was determined as parasite count per 200 white blood cells × 40 and was expressed as parasites per microlitre. The patients with Plasmodium falciparum were treated with amodiaquine plus S/P, and patients with P. vivax infection received chloroquine and primaquine.

Preparation and amplification of DNA

Parasite genomic DNA was prepared from dried bloodspots by the Chelex method as previously described (Kain & Lanar 1991). The obtained supernatant (approximately 100 μl), containing the isolated DNA, was transferred to a new tube and stored at −20 °C. We used a nested PCR method described by Duraisingh et al. (1998) and Jelinek et al. (1998) in order to detect polymorphisms on the DHFR and DHPS genes of all samples.

Restriction fragment length polymorphisms (RFLP)

The PCR product was incubated with restriction enzymes according to the manufacturer's instructions (New England Biolabs, Beverly, MA, USA). The product of the primer pair F-M4 was cut by AluI at codon 108 (serine). In the same PCR product, BstNI detected the 108-threonine mutation and XmnI detected 59-arginine. The amplified product of the M3-F/primer pair was cut by DraI to detect 164-leucine and by BsrI to detect 108-asparagine. A restriction site for NlaIII at this PCR product was destroyed by the mutation from 16-alanine to 16-valine and the restriction site for Tsp509I at 51-asparagine was destroyed by the mutation to 51-isoleucine. For the DHPS gene, the product of K–K/was cut by MnlI to identify 436-serine, MSPAI to detect 436-alanine, AvaII for 437-glycine, MwoI for 437-alanine and FokI for 540-glutamic acid. The product of L–L/was cut by BstUI to detect 581-alanine, Bsl1 for 581-glycine, MwoI for 613-alanine and AgeI for 613-threonine or BsaWI for 613-threonine and serine. DNA from established laboratory strains of P. falciparum (K1, FCR3, 3D7) served as controls of PCR and enzyme digests.


Thirty-six samples of dried bloodspots from Colombia were examined by nested PCR. Twenty-five (69.4%) of the 36 samples tested positive for a P. falciparum infection. The other samples tested positive for vivax malaria. DNA from all positive samples was amplified by nested PCR systems according to specific primer pairs as described above. Nineteen samples showed the expected bands at 522 bp (M3-F/) as well as the 326 bp (M4-F) for the DHFR gene (76%). The DHPS specific bands for the primer pair L–L/at 161 bp were presented by 18 (72%) samples, whereas only 13 (52%) samples showed the specific 438 bp bands for the K–K/primer pair. After digestion with the specific restriction enzymes, we found that mutations occurred in four of the 10 examined codons. Mutations were detected in three codons of the DHFR gene: codon 51, 59 and 108, while only codon 437 of the DHPS gene was mutated in some samples (Table 1). Four samples (21%) were digested by Tsp509I demonstrating a 51-asparagine to 51-isoleucine shift for the DHFR gene. A similar result was found for the XmnI digestion: three isolates (16%) showed a 59-cysteine to 59-arginine mutation. In both primer pairs, M3-F/as well as M4-F, the digestion with the enzymes BsrI and AluI showed 12 mutations, 108-serine to 108-asparagine (63%). No mutation occurred in codons 16 and 164. Concerning the DHPS gene, mutations were only found at codon 437. Three samples tested positive for the 437-glycine mutation (23%). No mutations were found in the other remaining codons of the DHPS gene (Table 1).

Table 1.  Polymorphisms in the codons of the DHFR and DHPS genes in P. falciparum isolates from Colombia
GeneCodonResults (%)
  1. Positions where no mutations occurred are not shown in the Table.

DHFR16-alanine19 (100)
51-asparagine15 (79)
51-isoleucine 4 (21)
59-cysteine16 (84)
59-arginine 3 (16)
108-serine 7 (37)
108-asparagine12 (63)
164-isoleucine19 (100)
DHPS436-serine13 (100)
437-alanine10 (77)
437-glycine 3 (23)
540-lysine13 (100)
581-alanine18 (100)
613-alanine18 (100)


Chloroquine-resistant malaria is often treated with the combination of S/P. With its widespread use in endemic areas, its effectiveness is becoming seriously compromised by the development of resistant strains. Evidence of correlation between polymorphisms in the codons of DHFR and DHPS genes of P. falciparum and in vitro resistance to antifolate drug combinations has been demonstrated in several studies (Foote et al. 1990; Peterson et al. 1990; Brooks et al. 1994; Triglia & Cowman 1994; Basco et al. 1995; Reeder et al. 1996). Understanding the molecular basis of resistance and its translation into methods of surveillance may help to monitor the process and spread resistance and could thus be the base of control programmes, and understanding the molecular basis of resistance and its translation into methods of surveillance could be essential tools for monitoring its spread in endemic areas (Eberl et al. 2001).

Examination of P. falciparum isolates from Apartadó (northern Colombia) revealed a low overall prevalence of mutations in the DHFR and DHPS genes (Table 1). Only four of the 10 studied codons were affected. In the DHPS gene, only codon 437 was mutated in some isolates. Although a mutation in codon 437 can contribute to the resistance to sulphadoxine (Wang et al. 1995; Cortese & Plowe 1998; Jelinek et al. 1998), most of the codons which may cause sulphadoxine resistance (codon 436, 581, 540 and 613) in the DHPS gene did not show any sings of mutations. Especially, 436-alanine, 436-phenylalanine and 613-serine seem to play a major role in the development of in vivo resistance against sulphadoxine (Foote et al. 1990; Peterson et al. 1990; Brooks et al. 1994; Basco & Ringwald 1998b).

Compared with the DHPS gene, the DHFR gene showed a larger number of mutations among the investigated isolates from Colombia. Three of the five examined DHFR codons were affected (Table 1). In codon 108, asparagine mutations occurred in 63%. In consideration of the importance of an 108-asparagine mutation in the DHFR gene for the development of resistance against pyrimethamine (Curtis et al. 1996; Sirawaraporn et al. 1997; Basco & Ringwald 1998a), these numbers indicate developing potential for antifolate resistance in Colombia. Mutations were found in two additional codons of the DHFR gene: 21% of the isolate samples had 51-isoleucine and 16% had 59-arginine. Both 51-isoleucine and 59-arginine increase the effect of 108-asparagine on pyrimethamine resistance (Curtis et al. 1996; Sirawaraporn et al. 1997; Basco & Ringwald 1998a).

Unlike previous studies from areas of low endemicity in West Africa (Eberl et al. 2001), we found a low overall prevalence of point mutations in the DHPS gene of P. falciparum in Colombian isolates, while mutations in the DHFR gene presented at a comparatively high rate. This developing potential of antifolate resistance should be taken into consideration when using S/P.