This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Marks, F.
Right arrow Articles by May, J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Marks, F.
Right arrow Articles by May, J.

 Previous Article  |  Next Article 

Antimicrobial Agents and Chemotherapy, February 2004, p. 466-472, Vol. 48, No. 2
0066-4804/04/$08.00+0     DOI: 10.1128/AAC.48.2.466-472.2004
Copyright © 2004, American Society for Microbiology. All Rights Reserved.

Genotyping of Plasmodium falciparum Pyrimethamine Resistance by Matrix-Assisted Laser Desorption-Ionization Time-of-Flight Mass Spectrometry

Florian Marks,1 Christian G. Meyer,1 Jürgen Sievertsen,1 Christian Timmann,1 Jennifer Evans,2 Rolf D. Horstmann,1 and Jürgen May1*

Bernhard Nocht Institute for Tropical Medicine, Hamburg, Germany,1 Kumasi Centre for Collaborative Research in Tropical Medicine, Kumasi, Ghana2

Received 26 August 2003/ Returned for modification 9 October 2003/ Accepted 24 October 2003


arrow
ABSTRACT
 
Increasing resistance, recrudescences, and treatment failure have led to the replacement of chloroquine with the combination of pyrimethamine (PYR) and sulfadoxine (SDX) as the first-line antimalarial drugs for treatment of uncomplicated Plasmodium falciparum malaria in several areas where this disease is endemic. The development of resistance to PYR-SDX is favored by incomplete treatment courses or by subtherapeutic levels in plasma. PYR-SDX resistance has been associated with several single-nucleotide polymorphisms (SNPs) in the P. falciparum dihydrofolate reductase (pfdhfr) and the P. falciparum dihydropteroate synthetase (pfdhps) genes. We have established assays based on matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) that conveniently allow the identification of SNPs associated with PYR resistance. Variants occurring at codon positions 16, 51, 59, and 108 of the pfdhfr gene were analyzed by MALDI-TOF MS in synthetic oligonucleotides to determine the detection threshold. In addition, 63 blood samples from subjects with P. falciparum parasitemia of various degrees were analyzed. The results were compared to those obtained by DNA sequencing of the respective gene fragment. The results of MALDI-TOF MS and DNA sequencing were consistent in 40 samples. In 23 samples two or three pfdhfr variants were detected by MALDI-TOF assays, whereas DNA-sequencing revealed one variant only. Simultaneous detection of two different mutations by biplex assays was, in principle, feasible. As demonstrated by the example of PYR resistance, MALDI-TOF MS allows for rapid and automated high-throughput assessment of drug sensitivity in P. falciparum malaria.


arrow
INTRODUCTION
 
The combination of pyrimethamine (PYR) and sulfadoxine (SDX) is one of the first-line drug regimens used to treat uncomplicated Plasmodium falciparum malaria in areas where this infection is endemic. Wide-scale use of PYR-SDX and inappropriate treatment regimens with residual drug levels in plasma are considered to essentially contribute to the emergence of resistance against PYR-SDX, as currently observed in many areas where malaria is endemic (7, 8, 23).


arrow
PYR resistance.
 
Resistance to PYR is conferred by a nonsynonymous single-nucleotide polymorphism (SNP) at codon position 108 of the P. falciparum dihydrofolate reductase gene (pfdhfr; AGC->AAC, pfdhfrS108N [12, 21]). The resulting amino acid substitution Ser->Asn exerts distinct steric constraints at the active site of the enzyme (24). Asp->Ile substitutions at codon position 51 (AAT->ATT; pfdhfrN51I) and/or a Cys->Arg exchange at position 59 (TGT->CGT; pfdhfrC59R) appear to progressively enhance PYR resistance if they occur along with pfdhfrS108N (3, 18, 21). The simultaneous emergence of a Ser->Thr substitution at position 108 (AGC->ACC; pfdhfrS108T) in combination with an Ala->Val exchange at position 16 (GCA->GTA; pfdhfrA16V) within the same clone confers resistance to cycloguanil, while sensitivity to PYR is only moderately affected (13, 18, 21). This combination of SNPs occurs rarely and has only been found in a few countries thus far (i.e., The Gambia, Uganda, Colombia, Brazil, and Thailand [1, 13]).

pfdhfrS108N/N51I/C59R is the haplotype most strongly associated with PYR resistance in Africa (19, 20). Selection for the pfdhfrS108N genotype has been shown to be linked to treatment with PYR drugs (18) and, accordingly, the rise of resistance and a continuous decline of the efficacy of PYR-SDX has been attributed to increased PYR consumption (14, 18, 23). A compilation of genes involved in drug resistance against falciparum malaria has been published recently (11).

A common feature of P. falciparum malaria in areas where the disease is endemic is the occurrence of multiclonal infections (9, 10), with individuals frequently and simultaneously being infected with the wild type and with strains exhibiting different drug resistance phenotypes. The quality of assays for the determination of resistant strains is, therefore, dependent on their ability to detect resistant clones of low parasitemia in multiclonal P. falciparum infections. In general, standardized genotyping of multiclonal P. falciparum is an unsolved problem (5).

Because of its rapid emergence, PYR resistance should be monitored in areas of wide-scale use. In addition, the prediction of expected PYR efficiency may be useful in countries where its introduction as a first-line drug is considered (6).


arrow
MALDI-TOF MS.
 
Matrix-assisted laser desorption-ionization time-of-flight mass spectrometry (MALDI-TOF MS) provides a technique for analyzing unfragmented molecules of high masses (>1 million Da), including biomolecules such as proteins and nucleic acids (15). In addition to its application in proteomics, MALDI-TOF MS-based SNP analysis has attracted increasing interest in recent years as a technique for high-throughput analyses.

To date, MALDI-TOF MS has neither been applied widely in pharmacogenomics nor used in the analysis of genetic variability of parasites. However, in the near future SNP analyses may play an important role in pharmacogenomics, and MALDI-TOF MS might well develop into a standard technique, e.g., in the assessment of drug sensitivity (2, 4, 17).

The present study was conducted to assess the sensitivity of MALDI-TOF MS in the detection of parasitic mutant variants associated with PYR resistance in samples of human DNA with natural P. falciparum infections in comparison to standard DNA-sequencing. In addition, the validity and sensitivity of the assay was examined with a serial dilution of synthetic oligonucleotides exhibiting either the wild type or the mutant variant.

(This research was conducted by F. Marks in partial fulfillment of the requirements for a Ph.D. thesis at the Bernhard Nocht Institute for Tropical Medicine, Hamburg, Germany.)


arrow
MATERIALS AND METHODS
 
Study population. In the course of a longitudinal study on malariometric indices in an area holoendemic for P. falciparum malaria (Ashanti-Region, Ghana, West Africa), blood samples were collected from healthy adults. Sixty-three samples with proven P. falciparum parasitemia were used to screen for PYR-resistant strains. Parasitemia was determined microscopically and confirmed by PCR assays. Ethical clearance was obtained through the Ethics Committee of the School of Medical Sciences, Kwame Nkrumah University of Science and Technology, Kumasi, Ghana.

pfdhfr PCR. Genomic DNA was isolated from peripheral venous blood. Two PCR primers were designed to span the region exhibiting mutations relevant to PYR resistance and to yield a PCR product of 572 bp. Sequences of the oligonucleotides are given in Table 1. PCR was performed in a reaction volume of 50 µl containing primers (final concentration, 0.4 µM), deoxynucleoside triphosphates (dNTPs; 4x 62.5 µM), 1x Qiagen buffer, MgCl2 (0.5 mM), 1 U of Hotstar-Taq polymerase (Qiagen, Valencia, Calif.), and 80 ng of template DNA consisting of genomic human DNA with parasitic DNA concentrations corresponding to the individual parasite burdens. After an initial denaturation step (15 min for 95°C), 49 cycles of denaturation at 94°C for 30 s, primer annealing at 50°C for 40 s, and extension at 72°C for 1 min were run. Elongation of amplicons was eventually completed for 10 min at 72°C. The expected size and the quality of the PCR products was monitored on ethidium bromide-stained 1% agarose gels. A single PCR was performed on each sample only. The resulting product served as the template for both the MALDI-TOF MS and the DNA sequencing procedures.


View this table:
[in this window]
[in a new window]
  		TABLE 1. Oligonucleotides for PCR, primer extension reactions, and sensitivity assays

Purification of the PCR product. The wells of a Multiscreen filter plate (Millipore Corp., Bedford, Mass.) were filled with 100 mg of GR50 Superfine Sephadex (Amersham, Buckinghamshire, United Kingdom) and 300 µl of high-pressure liquid chromatography-grade water. After 3 h at ambient temperature, the plate was centrifuged (15 min, 912 x g) for removal of water and packing of the Sephadex columns within the wells. PCR products (50 µl) were loaded onto the columns and eluted (5 min of centrifugation at 912 x g) to remove the dNTPs, primers, and contaminants. The purification efficiency was monitored on agarose gels. Subsequently, the purified DNA fragments were subjected to DNA sequencing and MALDI-TOF MS.

DNA sequencing. DNA sequencing was performed on an ABI 3100 Sequencer (Applied Biosystems, Foster City, Calif.) and a BigDye terminator cycle sequencing ready reaction kit (v3.0; Applied Biosystems) according to the manufacturer's instructions and with primers identical to those of the PCR.

MALDI-TOF MS. In brief, SNP analysis by MALDI-TOF MS was performed as follows (15). After PCR amplification and purification of the DNA fragment, an oligonucleotide probe was annealed to the PCR product adjacent to the SNP under consideration, and a primer extension reaction was initiated. Elongation of the oligonucleotide probe depends on the known variability of the template (e.g., exerted through SNPs), which is reflected in the content and composition of stop nucleotides (ddNTPs) within the reaction mixture. The reaction product was purified and embedded into a matrix on an appropriate target for subsequent desorption/ionization. The intensity of peaks generated by one laser and acceleration impulse (the intensity is commonly given in arbitrary units) depends mainly on the concentration of ions that are desorbed and ionized with one laser impulse and on the preselected voltage of the acceleration impulse. The x axis of MALDI-TOF spectra that are generated represent increasing m/z ratios, and the y axis shows arbitrary units, indicating the relative abundance (i.e., the relative intensity) of each ion, which is related to the number of times ions of that particular m/z ratio strike the detector. The assignment of relative abundance depends on the most abundant ion, which is 100%. The intensities of all other ions are given as proportions of that most abundant ion and are expressed as percentages.

For MALDI-TOF MS, a primer extension reaction was performed for the mutations that are involved in conferring PYR resistance, namely, at codon positions 16, 51, 59, and 108 of the pfdhfr gene. For variants of pfdhfr16/pfdhfr108, a biplex reaction was established to simultaneously detect SNPs. The design and orientation of extension primers (sense or antisense primer) was dependent on the sequence adjacent to the SNP under question. Sequences of the oligonucleotide primers are given in Table 1.

Primer extension reactions (10 µl) were executed by using dNTPs of each base, excluding the equivalent of one of the two variants under question, which was a ddNTP base. Thus, the reaction mixture consisted of each dCTP, dGTP, and dATP (200 µM) ddTTP (200 µM), an extension primer (0.75 mM), 1x Thermo-Sequenase buffer, 1 U of Thermo-Sequenase (Amersham), and purified PCR product (ca. 20 ng) in H2O for reactions to detect pfdhfr51 and pfdhfr59 variants. For the reactions established to identify pfdhfr16 and pfdhfr108 variants, ddATP was used instead of dATP.

For the pfdhfr16 and pfdhfr108 primer extension reactions, an initial denaturation was performed at 94°C for 1 min, followed by 49 cycles of denaturation at 94°C for 15 s, annealing at 54°C for 1 min, and extension and/or elongation at 72°C for 30 s. For pfdhfr51and pfdhfr59 reactions, the annealing temperature was 52°C in 49 cycles. After the extension reaction, the single-stranded primer extension products were purified either manually or automatically (Genesis Workstation 200; Tecan, Durham, N.C.) by using a Magnetic Bead DNA Purification Kit 1000 (GenoPure Oligo; Bruker Daltonics, Bremen, Germany).

Then, 1 µl of the matrix (3-hydroxypicolinic-acid [3-HPA] at 5 g/liter in 0.7 g of diammonium citrate/liter) was evenly applied onto a MALDI-TOF metal 400-µm anchor plate with hydrophilic anchors in a hydrophobic surface (Bruker Daltonics). After the matrix was dried (ca. 15 min), 1 µl of the analyte was spotted onto the matrix to cocrystallize with the 3-HPA.

MALDI-TOF sensitivity assay. In order to determine the sensitivity and validity of the MALDI-TOF MS assay for the various SNPs and to compare it with results obtained from human genomic DNA samples, oligonucleotides exhibiting the pfdhfr51, pfdhfr59, and pfdhfr108 variants were synthesized and used as templates (sequences of oliogonucleotides given in Table 1). For each mutation, a sequential dilution series, containing both the wild-type and mutant oligonucleotide was prepared (0, 5, 9, 17, 33, 50, 66, 83, 91, 95, and 100%). The summarized percentage of both olignucleotides was always 100%. We found that 1% was equivalent to 2 pmol of synthetic DNA of one of the two variants. Primer extension, purification, and MALDI-TOF MS were performed as described above.

Statistics. McNemar's {chi}2 test was calculated to estimate differences between results obtained through MALDI-TOF MS and DNA sequencing. This nonparametric test assesses the significance of differences between two dependent samples. Significant results imply that frequencies or proportions of discordant pairs are heterogeneous.

Results of the comparisons between the determination of variability by MALDI-TOF MS and by DNA sequencing were transformed to fit a 2x2 classification table (biclonal versus monoclonal infection). Discordant results were used to estimate differences between the two methods. Discordance in this context is defined as the presence of more than one allele detected in one of the two assays but not in the other (discordant results are indicated in boldface in Fig. 4).



View larger version (37K):
[in this window]
[in a new window]
  		FIG. 4. Sensitivity and specificity of MALDI-TOF MS typing compared to DNA sequencing. (A) pfdhfr51 ({chi}2 [McNemar] = 8.1; P < 0.005); (B) pfdhfr59 ({chi}2 [McNemar] = 17.1; P < 0.0001); (C) pfdhfr108 ({chi}2 [McNemar] = 5.1; P < 0.03). The nucleotide(s) in question are indicated in boldface (i.e., the wild type/mutant variant [wt/mut]); numbers in boldface indicate the number of samples with discordant results.


arrow
RESULTS
 
Sixty-three blood samples with proven P. falciparum parasitemia collected in an area holoendemic for malaria in Ghana, West Africa, were screened for four SNPs of the pfdhfr gene associated with PYR resistance. Parasitemia was confirmed, and its degree was assessed by standard microscopy of Giemsa-stained blood smears.

The pfdhfr wild-type sequence and its allelic variants pfdhfrN51I, pfdhfrC59R, and pfdhfrS108N could readily and reproducibly be detected by MALDI-TOF MS according to differences in their times of flight, given as the ratio of mass to the number of charges of the analyte ion (m/z). The variant pfdhfrA16V was not detected in our study sample. As a reference, automated DNA sequencing was performed on all samples.

Sensitivity assays. To assess the detection limits and to determine the correlation of true allele ratios with ratios measured by MALDI-TOF MS, the arbitrary units were transformed to relative values (relative abundancies). The background noise in our assays was between 150 and 500 arbitrary units, which corresponds to an overall detection limit at a relative abundance of ca. 10%. The true allele ratios mixed from synthetic oligonucleotides and the observed allele ratios as measured by MALDI-TOF MS were in strong linear correlation (r2 for all variants of >0.95; P < 0.001; Fig. 1). Figure 1 shows the results from eight independent measurements for each of the variant mixtures. The detection thresholds varied, depending on the mutation under question, between proportions of 5 and 17%, a level corresponding to 10 to 34 pmol, of the respective oligonucleotide: pfdhfr51(A), 5%; pfdhfr51(T), 9%; pfdhfr59(T), 17%; pfdhfr59(C), 5%; pfdhfr108(G), 17%; and pfdhfr108(A), 9%. Intercepts indicate that the proportions of the variants pfdhfr51(A), pfdhfr59(C), and pfdhfr108(A) were overestimated (Fig. 1A to C, respectively).



View larger version (21K):
[in this window]
[in a new window]
  		FIG. 1. Sensitivity assays performed in a series of eight observations, each with synthetic oligonucleotides for the determination of the limit of pfdhfr SNP detection. Oligonucleotides exhibiting the wild-type sequence were mixed with those exhibiting the mutant variant in dilution series with defined proportions of oligonucleotides; a concentration of 1% corresponds to 2 pmol of synthetic DNA. Arbitrary units were transformed to percent relative abundancies. Due to the background noise and contaminating electrolytes, the detection threshold was 10% of the relative abundance. (A) r2 = 0.98, P < 0.001, slope 0.889, detection limits for pfdhr51(T) of 9% and for pfdhr51(A) of 5%; (B) r2 = 0.96, P < 0.001, slope 0.839, detection limits for pfdhr59(C) of 5% and for pfdhr59(T) of 17%; (C) r2 = 0.96, P < 0.001, slope 0.931, detection limits for pfdhr108(A) of 9% and for pfdhr108(G) of 17%.

MALDI-TOF spectra. MALDI-TOF spectra and corresponding chromatograms of DNA-sequencing, representing the variants identified in our study group, are shown in Fig. 2.



View larger version (29K):
[in this window]
[in a new window]
  		FIG. 2. MALDI-TOF spectra of the determination of variability at codon position 51 (A), position 59 (B), and position 108 (C) of the pfdhfr gene. The corresponding results of DNA sequencing are shown in insets. The intensity is given in arbitrary units that depend on the intensity of the highest peak. With an m/z of 22 of sodium, contaminations exhibiting the expected m/z with one or two sodium ions can clearly be discriminated.

Figure 2A shows the spectrum obtained for the determination of pfdhfr51 variants. The peak at m/z 5,512 corresponds to the 18mer primer for the extension reaction, and the peaks at m/z values of 5,800 and 6,114 represent the mutant and the wild-type variant, respectively. The corresponding DNA sequence (Fig. 2A, inset) confirms the presence of an A->T substitution at position 51 of the pfdhfr gene. Figure 2B presents the results of the MALDI-TOF MS assay designed for determining variability at position 59 of the pfdhfr gene and the corresponding DNA sequence (primer for the extension reaction not shown). Figure 2C shows a peak for the extension primer (m/z 6,229) and peaks for the mutant and the wild-type variant at m/z 6,526 and m/z 7,135, respectively, a result again consistent with DNA sequencing.

DNA has a strong tendency to form concatemers with potassium and sodium ions, eventually resulting in a distinguishable increase of masses according to the m/z values of adducts (Fig. 2 and 3). Such concatemers are generally considered the limiting factors in analysis of DNA fragments of >40 bp.



View larger version (28K):
[in this window]
[in a new window]
  		FIG. 3. Mass spectrum of a biplexed assay for the detection of SNPs at codon positions 16 and 108 of the pfdhfr gene. The intensity is given in arbitrary units. For position 16, the wild-type variant was detected only. Variability at position 16 of pfdhfr is rare and did not occur among the study population.

The intensity of the two peaks allows a rough estimation of the ratio of the wild-type to the mutant variants. However, in order to obtain more valid information on the proportion of variants within one sample, series of preparations and series of measurements from each preparation must be performed.

Biplex assays. Multiplexed MALDI-TOF MS reactions turned out to be limited by constraints exerted through either distinct properties of the DNA sequences adjacent to the SNPs or through weakness of signals, not allowing unambiguous discrimination from background noise or contamination peaks. Biplexed detection was feasible only for the combined identification of the SNPs at codon positions 16 and 108 (Fig. 3). Since the mutant variant pfdhfr16 was not detected in our study group, one peak only resulted for the nucleotide at codon position 16 (m/z 5,699; wild-type strain; Fig. 3). The assay was designed to yield a peak at m/z 5,081 for the mutant variant.

Comparison of MALDI-TOF MS versus DNA sequencing results. A comparison of results obtained by MALDI-TOF MS with those of DNA sequencing for variability at codon positions 51, 59, and 108 of the pfdhfr gene is presented in Fig. 4. Concordant results of both techniques were obtained in 53, 43, and 56 samples for the codon positions 51, 59, and 108, respectively. MALDI-TOF MS detected mixed infections in an additional 10, 19, and 7 samples at the respective codon positions (discordant results are given in boldface [pfdhfr51, PMcNemar < 0.005], pfdhfr59, PMcNemar < 0.0001, and pfdhfr108, PMcNemar < 0.03]). Notably, several DNA samples exhibited more than one of the mutations under consideration, resulting in these 36 discordant results. No mixed infections were identified by DNA sequencing that were not detected by MALDI-TOF MS. The discordant results unambiguously indicate a higher sensitivity of MALDI-TOF MS in identifying more than one allelic variant.

The 36 discordant results were observed in 23 of the 63 human DNA samples. Of these 23 samples, the proportion of samples with discordant results for more than one of the positions was high. In 15 samples in which one variant was only identified by sequencing at codon positions 51, 59, and 108, an additional variant was detected by MALDI-TOF MS. In five samples, MALDI-TOF MS detected two mutations at different positions, while DNA sequencing identified one variant. In three samples assigned one variant by DNA sequencing, two variants were detectable at all three positions by MALDI-TOF MS.


arrow
DISCUSSION
 
This study was designed to establish MALDI-TOF MS-based assays for the determination of mutant variants in the P. falciparum dihydrofolate reductase gene (pfdhfr) that are relevant to resistance against PYR. The test was performed first as a sensitivity assay on dilution series of synthetic 42- to 45mer oligonucleotides with defined amounts of DNA and, second, on 63 clinically healthy, but P. falciparum-infected adult individuals from an area holoendemic for the transmission of P. falciparum malaria in Ghana, West Africa. The results of MALDI-TOF MS analyses were compared to those of automated DNA sequencing.

When we considered a background noise of ca. 10% of the relative abundance in the MALDI-TOF MS spectra, the variants were detectable at concentrations of 2.5 to 15%, corresponding to 5 to 30 pmol of synthetic DNA. This result is in agreement with a study on the determination of SNP frequencies in DNA pools by MS in which the detection limit of the minor allele was determined to be 5 to 10% (19).

The results indicated differences in the sensitivity for detecting either the wild-type variant (codon position 51 of the pfdhfr gene) or the mutant variants (positions 59 and 108). Therefore, with low concentrations of distinct DNA species (e.g., in our study low parasitemia of different P. falciparum clones), the frequency of the variant with the lower detection limit might be overestimated. It is known that the frequency of that allele, which in the MALDI-TOF assay corresponds to a lower mass fragment, is often overestimated. This bias is mostly a methodological consequence of (i) preference of ionization of smaller nucleotides, (ii) detector bias toward smaller DNA molecules, and (iii) increased formation of sodium adducts with larger oligonucleotides (22). Notably, a systematic deviation toward smaller molecules was not observed with the six variant DNA fragments examined here. Due to the strong linear correlation between the measured and the true allele ratios, a correction factor for that overestimation may be applied that takes the slope of the regression formula into account (Fig. 1).

The assays performed on human DNA samples of P. falciparum-infected, but clinically healthy adults showed that MALDI-TOF MS was more sensitive than DNA sequencing in the detection of infections exhibiting both the wild-type and the mutant variants. Whereas MALDI-TOF MS detects variant nucleotides individually at different m/z ratios, yielding independent signals for each variant, the two variants are indicated by superimposed peaks in DNA sequencing. This inherently bears the risk that at low concentrations of one variant, e.g., resulting from low parasitemia or differences of sequence-dependent PCR efficiency, the signal may be hidden by background noise.

In looking for a bivariate marker of resistance, a general problem is that a determination can be made only on the occurrence of these two markers but not on the number of clones that carry that particular marker. It is, in fact, virtually impossible to estimate the relative proportions of different clones. This problem does not apply to the detection of human DNA variability in heterozygous individuals, in which the proportion of two variants is always equal.

Notably, a feature of P. falciparum infections in many areas where this organism is endemic is the occurrence of multiclonal infections, which are characterized by a variety of mutations not only in the pfdhfr gene but also in other genes (9, 10).

In addition, with differing parasitemias of clones, those associated with low parasitemia and exhibiting distinct markers of interest may be below the threshold of detection. The quality and the sensitivity of an assay is, therefore, crucially dependent on its ability to detect variants at the lowest concentrations, which, on the other hand, may be present in different clones. The high frequency of multiclonal infections results in a high frequency of heterogeneity at codon positions 51, 59, and 108 identified by MALDI-TOF MS, mutations known to progressively enhance the PYR resistance of P. falciparum (12, 18).

Our findings show that MALDI-TOF MS can be used as a sensitive and accurate method for typing of parasite gene polymorphisms. Mutations at codon positions 51, 59, and 108 of the P. falciparum pfdhfr gene were reproducibly and consistently identified. No mutations were found at position pfdhfr16, a finding corresponding to the previous observation that variants at that position are rare (1, 13). A biplexed assay for the detection of more than one variant was feasible only for the simultaneous detection of pfdhfr16 and pfdhfr108 variants. Involvement of one of the other mutations was restricted due to constraints exerted either by properties of the sequences not allowing for the identification of appropriate termination nucleotides for the extension reaction or by weak signals that could not be distinguished from background noise or sodium adducts.

MALDI-TOF MS appears to be superior to other techniques, such as restriction fragment length polymorphism assays, mutation specific-PCR, and oligonucleotide dot blot hybridization assays, that are commonly applied to determine pfdhfr variants (16). These methods are either time-consuming, costly, or unsatisfactory with regard to sensitivity and/or specificity. Limitations that apply to all methods, including MALDI-TOF MS, include the lack of internal controls, leading to the possibility that a clone of low parasitemia may not be detected, and the fact that only known variants can be detected.

Technical drawbacks of MALDI-TOF MS have recently been described in detail (15). These apply in particular to the technique itself and to the preparation of samples. Purification of the PCR products that are subjected to MALDI-TOF MS requires special attention and is considered a crucial step.

Standard MALDI-TOF MS allows the use of targets with 384 or 1,536 wells, facilitating simultaneous preparation steps and, in principle, high-throughput genotyping of DNA species. This applies in particular if DNA preparation steps are performed in an automated fashion. However, MALDI-TOF MS identification is limited to analyses of small DNA fragments and is currently restricted by financial constraints due to the high cost of the necessary sophisticated hardware and reagents. Recent advances in the development of portable and, hopefully affordable, MS devices may allow the future use of this sensitive tool for the detection of microbial drug resistance genes in areas that are subjected to the highest burden of infectious diseases.


arrow
ACKNOWLEDGMENTS
 
This study was supported by a grant from the Bundesministerium für Bildung und Forschung (01KA0292) and by the German Human Genome Project (O1KW9918).


arrow
FOOTNOTES
 
* Corresponding author. Mailing address: Bernhard Nocht Institute for Tropical Medicine, Department of Molecular Medicine, Bernhard-Nocht-Str. 74, 20359 Hamburg, Germany. Phone: 49-40-42818-369. Fax: 49-40-42818-512. E-mail: may{at}bni.uni-hamburg.de. Back


arrow
REFERENCES
 
    1
  1. Biswas, S. 2001. Plasmodium falciparum dihydrofolate reductase Val-16 and Thr-108 mutation associated with in vivo resistance to antifolate drug: a case study. Indian J. Malariol. 38:76-83.[Medline]
    2. 2
  2. Brennan, M. D. 2001. High throughput genotyping technologies for pharmacogenomics. Am. J. Pharmacogenomics 1:295-302.[CrossRef][Medline]
    3. 3
  3. Cowman, A. F. 1997. The mechanisms of drug action and resistance in malaria. Mol. Genet. Drug Res. 3:221-246.
    4. 4
  4. Evans, W. E., and M. V. Relling. 1999. Pharmacogenomics: translating functional genomics into rational therapeutics. Science 286:487-491.[Abstract/Free Full Text]
    5. 5
  5. Farnert, A., A. P. Arez, H. A. Babiker, H. P. Beck, A. Benito, A. Bjorkman, M. C. Bruce, D. J. Conway, K. P. Day, L. Henning, O. Mercereau-Puijalon, L. C. Ranford-Cartwright, J. M. Rubio, G. Snounou, D. Walliker, J. Zwetyenga, and V. E. do Rosario. 2001. Genotyping of Plasmodium falciparum infections by PCR: a comparative multicentre study. Trans. R. Soc. Trop. Med. Hyg. 95:225-232.[CrossRef][Medline]
    6. 6
  6. Kublin, J. G., R. S. Witzig, A. H. Shankar, J. Q. Zurita, R. H. Gilman, J. A. Guarda, J. F. Cortese, and C. V. Plowe. 1998. Molecular assays for surveillance of antifolate-resistant malaria. Lancet 351:1629-1630.[CrossRef][Medline]
    7. 7
  7. Le Bras, J., and R. Durand. 2003. The mechanisms of resistance to antimalarial drugs in Plasmodium falciparum. Fundam. Clin. Pharmacol. 17:147-153.[CrossRef][Medline]
    8. 8
  8. May, J., and C. G. Meyer. 2003. Chemoresistance in falciparum malaria. Trends Parasitol. 19:432-435.
    9. 9
  9. May, J., and C. G. Meyer. 2003. Association of Plasmodium falciparum chloroquine resistance transporter variant T76 with age-related plasma chloroquine levels. Am. J. Trop. Med. Hyg. 68:143-146.[Abstract/Free Full Text]
    10. 10
  10. May, J., F. P. Mockenhaupt, O. G. Ademowo, A. G. Falusi, P. E. Olumese, U. Bienzle, and C. G. Meyer. 1999. High rate of mixed and subpatent malarial infections in southwest Nigeria. Am. J. Trop. Med. Hyg. 61:339-343.[Abstract]
    11. 11
  11. Meyer, C. G., J. May, A. P. Arez, J. P. Gil, and V. do Rosario. 2002. Genetic diversity of Plasmodium falciparum: asexual stages. Trop. Med. Int. Health 7:395-408.[CrossRef][Medline]
    12. 12
  12. Peterson, D. S., D. Walliker, and T. E. Wellems. 1988. Evidence that a point mutation in dihydrofolate reductase-thymidylate synthase confers resistance to pyrimethamine in falciparum malaria. Proc. Natl. Acad. Sci. USA 85:9114-9118.[Abstract/Free Full Text]
    13. 13
  13. Peterson, D. S., W. K. Milhous, and T. E. Wellems. 1990. Molecular basis of differential resistance to cycloguanil and pyrimethamine in Plasmodium falciparum malaria. Proc. Natl. Acad. Sci. USA 87:3018-3022.[Abstract/Free Full Text]
    14. 14
  14. Plowe, C. V., A. Djimde, T. E. Wellems, S. Diop, B. Kouriba, and O. K. Doumbo. 1996. Community pyrimethamine-sulfadoxine use and prevalence of resistant Plasmodium falciparum genotypes in Mali: a model for deterring resistance. Am. J. Trop. Med. Hyg. 55:467-471.
    15. 15
  15. Pusch, W., J. H. Wurmbach, H. Thiele, and M. Kostrzewa. 2002. MALDI-TOF mass spectrometry-based SNP genotyping. Pharmacogenomics 3:537-548.[CrossRef][Medline]
    16. 16
  16. Ranford-Cartwright, L. C., K. L. Johnston, A. M. Abdel-Muhsin, B. K. Khan, and H. A. Babiker. 2002. Critical comparison of molecular genotyping methods for detection of drug-resistant Plasmodium falciparum. Trans. R. Soc. Trop. Med. Hyg. 96:568-572.
    17. 17
  17. Shi, M. M. 2002. Technologies for individual genotyping: detection of genetic polymorphisms in drug targets and disease genes. Am. J. Pharmacogenomics 2:197-205.[CrossRef][Medline]
    18. 18
  18. Sibley, C. H., J. E. Hyde, P. F. Sims, C. V. Plowe, J. G. Kublin, E. K. Mberu, A. F. Cowman, P. A. Winstanley, W. M. Watkins, and A. M. Nzila. 2001. Pyrimethamine-sulfadoxine resistance in Plasmodium falciparum: what next? Trends Parasitol. 17:582-588.[CrossRef][Medline]
    19. 19
  19. Wang, P., C. S. Lee, R. Bayoumi, A. Djimde, O. Doumbo, G. Swedberg, L. D. Dao, H. Mshinda, M. Tanner, W. M. Watkins, P. F. Sims, and J. E. Hyde. 1997. 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. 89:161-177.[CrossRef][Medline]
    20. 20
  20. Warhurst, D. C. 1997. Drug-resistant malaria: laboratory and field investigations. J. Pharm. Pharmacol. 2:3-7.
    21. 21
  21. Warhurst, D. C. 2002. Resistance to antifolates in Plasmodium falciparum, the causative agent of tropical malaria. Sci. Prog. 85:89-111.[Medline]
    22. 22
  22. Werber, M., M. Sych, N. Herbon, T. Illig, I. R. König, and M. Wjist. 2002. Large-scale determination of SNP allele frequencies in DNA pools using MALDI-TOF mass spectrometry. Hum. Mut. 20:57-64.[CrossRef][Medline]
    23. 23
  23. World Health Organization. 2001. Antimalarial drug combination therapy. Report of a WHO Technical Consultation. W. H. O. Tech. Rep. Ser. 2001:1-36.
    24. 24
  24. Yuvaniyama, J., P. Chitnumsub, S. Kamchonwongpaisan, J. Vanichtanankul, W. Sirawaraporn, P. Taylor, M. D. Walkinshaw, and Y. Yuthavong. 2003. Insights into antifolate resistance from malarial DHFR-TS structures. Nat. Struct. Biol. 10:357-365.[CrossRef][Medline]

Antimicrobial Agents and Chemotherapy, February 2004, p. 466-472, Vol. 48, No. 2
0066-4804/04/$08.00+0     DOI: 10.1128/AAC.48.2.466-472.2004
Copyright © 2004, American Society for Microbiology. All Rights Reserved.



This article has been cited by other articles:

  • Marks, F., Evans, J., Meyer, C. G., Browne, E. N., Flessner, C., von Kalckreuth, V., Eggelte, T. A., Horstmann, R. D., May, J. (2005). High Prevalence of Markers for Sulfadoxine and Pyrimethamine Resistance in Plasmodium falciparum in the Absence of Drug Pressure in the Ashanti Region of Ghana. Antimicrob. Agents Chemother. 49: 1101-1105 [Abstract] [Full Text]  

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowReprints and Permissions
Right arrow Copyright Information
Right arrow Books from ASM Press
Right arrow MicrobeWorld
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Marks, F.
Right arrow Articles by May, J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Marks, F.
Right arrow Articles by May, J.

Copyright © 2009 by the American Society for Microbiology.   For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»