Transgenic Plasmodium parasites stably expressing Plasmodium vivax dihydrofolate reductase-thymidylate synthase as in vitro and in vivo models for antifolate screening

Background Plasmodium vivax is the most prevalent cause of human malaria in tropical regions outside the African continent. The lack of a routine continuous in vitro culture of this parasite makes it difficult to develop specific drugs for this disease. To facilitate the development of anti-P. vivax drugs, bacterial and yeast surrogate models expressing the validated P. vivax target dihydrofolate reductase-thymidylate synthase (DHFR-TS) have been generated; however, they can only be used as primary screening models because of significant differences in enzyme expression level and in vivo drug metabolism between the surrogate models and P. vivax parasites. Methods Plasmodium falciparum and Plasmodium berghei parasites were transfected with DNA constructs bearing P. vivax dhfr-ts pyrimethamine sensitive (wild-type) and pyrimethamine resistant (mutant) alleles. Double crossover homologous recombination was used to replace the endogenous dhfr-ts of P. falciparum and P. berghei parasites with P. vivax homologous genes. The integration of Pvdhfr-ts genes via allelic replacement was verified by Southern analysis and the transgenic parasites lines validated as models by standard drug screening assays. Results Transgenic P. falciparum and P. berghei lines stably expressing PvDHFR-TS replacing the endogenous parasite DHFR-TS were obtained. Anti-malarial drug screening assays showed that transgenic parasites expressing wild-type PvDHFR-TS were pyrimethamine-sensitive, whereas transgenic parasites expressing mutant PvDHFR-TS were pyrimethamine-resistant. The growth and sensitivity to other types of anti-malarial drugs in the transgenic parasites were otherwise indistinguishable from the parental parasites. Conclusion With the permanent integration of Pvdhfr-ts gene in the genome, the transgenic Plasmodium lines expressing PvDHFR-TS are genetically stable and will be useful for screening anti-P. vivax compounds targeting PvDHFR-TS. A similar approach could be used to generate transgenic models specific for other targets of interest, thus facilitating the development of anti-P. vivax drugs in general.


Background
Current anti-malarial drug development efforts are focused on screening for lead compounds against Plasmodium falciparum, the most lethal species. Plasmodium falciparum can be routinely cultured in vitro and is thus amenable to high throughput compound screening. Anti-malarial drug development against other human Plasmodium species, particularly Plasmodium vivax -which is the most prevalent cause of human malaria in tropical regions, with an estimated 80 million cases annually [1], is neglected in comparison, since continuous in vitro culture methods are not available for these species. The morbidity and mortality caused by P. vivax infection are greater than previously believed [2]. Chloroquine has been used as the standard treatment for blood stage vivax malaria for more than 40 years; however, chloroquine-resistant P. vivax has been reported in many parts of the world [3][4][5]. Therefore, there is an urgent need for new anti-vivax anti-malarial drugs.
One of the validated drug targets for the treatment of malaria infection is dihydrofolate reductase (DHFR; EC1.5.1.3), an essential enzyme for folate biosynthesis [6,7]. The efficacy of current anti-folate drugs targeting Plasmodium DHFR is compromised by mutations in the dhfr gene, which confer different levels of resistance to these drugs [8][9][10]. Point mutations in the P. vivax dhfr gene equivalent to antifolate resistance mutations found in P. falciparum have been associated with antifolate resistance in P. vivax in vitro [11][12][13], leading to the conclusion that wild-type P. vivax is sensitive to antifolates, and resistance develops through dhfr mutations, similar to the case in P. falciparum.
Despite the emergence of drug-resistant dhfr mutants, DHFR-TS is still an attractive target for anti-malarial drug development owing to the availability of targetbased screening models and crystal structures of both P. falciparum [14] and P. vivax DHFR enzymes [15]. However, unlike P. falciparum, the development of antifolates directed against P. vivax has been hampered by the lack of a continuous in vitro parasite culture system. Clinical isolates can only be cultured for a short period of time [16,17]. Furthermore, mixed infection with other human malaria parasites can also complicate the drug screening results of clinical isolates. Recently, a P. vivax in vitro culture system using erythroblasts has been developed [18], but it is not practical for use in routine drug screening assay.
Surrogate cell expression systems expressing PvDHFR enzymes in yeast [19] and bacteria [20] have been developed as alternatives to parasite drug screening. However, these surrogate systems are of limited use since the level of target enzyme expression and mechanisms of drug metabolism differ markedly from Plasmodium parasites. A physiologically similar Plasmodium species is clearly a more attractive P. vivax surrogate, and this has been proven using transfection technology in which P. falciparum parasites expressing mutant PvDHFR-TS have been shown to recapitulate the mutant PvDHFR-TS antifolate resistance [21]. In the same manner, transgenic P. falciparum parasites episomally expressing drug resistant PvDHFR-TS mutant enzymes have also been developed for studying P. vivax drug resistance. However, these transgenic P. falciparum parasites still express endogenous PfDHFR-TS as well as episomally expressed PvDHFR-TS enzymes [22], which could confound drug screening results.
In order to facilitate antifolate screening specifically against PvDHFR-TS, transgenic P. falciparum and Plasmodium berghei parasites stably expressing PvDHFR-TS enzyme replacing the endogenous Pf or PbDHFR-TS were generated. These transgenic lines were evaluated as in vitro and in vivo Plasmodium models for direct assessment of the antifolate efficacy against PvDHFR-TS. These Plasmodium surrogate models were thus validated as alternative tools for screening PvDHFR-TS-targeted compounds.

Parasites
Plasmodium falciparum strains TM4/8.2 (wild-type DHFR) and K1CB1 (mutant DHFR at residues C59R +S108N) were gifts from Sodsri Thaithong, Faculty of Science, Chulalongkorn University, Thailand. Pyrimethamine-resistant strain K1CB1 was used as parental parasite to generate an in vitro P. vivax screening model. The parasites were maintained continuously in human erythrocytes at 37°C under 5% CO 2 , 1% O 2 and 94% N 2 in RPMI-1640 culture media supplemented with 24 mM NaHCO 3 , 40 μg/mL gentamicin, 25 mM HEPES and 10% human serum [23]. For generation of in vivo screening models, the transgenic P. berghei parasite line MRA-867 (PbGFP), a gift from Chris Janse and Andy Waters at Leiden University Medical Center, the Netherlands, was used. This line contains the gfp gene stably integrated as a single copy by double cross-over recombination into the 230p locus, and does not contain a drug selectable marker gene [24,25]. Frozen parasites from stock were mechanically passaged at least once through female BALB/c mice before experiments. Animals were bled from the heart and parasitaemia measured by Giemsa-stained thin blood smear. Infected erythrocytes were suspended in PBS and infection for experiments was carried out by intra-peritoneal injection of approximately 1 × 10 7 parasitized erythrocytes.

Experimental animals
Pathogen-free, six-week-old female BALB/c mice weighing 20-30 g were obtained from the National Laboratory Animal Center, Mahidol University, Thailand. They were kept for at least one week with tap water and pellet diet (CP diet 082, Perfect Companion Company, Bangkok, Thailand) ad libitum at 22-25°C and a 12-hour light/12-hour dark cycle. Experiments were started in seven-to eight-week-old animals. Animal experiments were ratified by the Ethical Committee on Animal Experimentation, Faculty of Medicine, Chiang Mai University, Chiang Mai, Thailand and by the Ethical Committee on Animal Experimentation, National Center for Genetic Engineering and Biotechnology (BIOTEC), Thailand. Animal experiments conformed to international and national guidelines for ethical conducts on the care and humane use of animals.

Construction of transfection plasmids
Wild-type and mutant P. vivax dhfr-ts genes were kindly given by Ubolsree Leartsakulpanich, BIOTEC, Thailand [11]. For generation of transgenic P. falciparum parasite stably expressing PvDHFR-TS, a transfection plasmid was constructed containing three expression cassettes: 1) blasticidin-S deaminase (bsd) as a positive selection marker under the control of 5' flanking region of P. falciparum camodulin gene (Pfcam) and 3' UTR of P. falciparum histidine rich protein 2 (Pfhrp2) [26], 2) A fusion gene of cytosine deaminase and uracil phosphoribosyl transferase (yfcu) from Saccharomyces cerevisiae as a negative selection marker under the control of 5' flanking region of P. falciparum heat shock protein 86 gene (Pfhsp86) and 3' UTR of P. berghei dihydrofolate reductase (Pbdhfr) [27], and 3) The wild-type dhfr-ts gene of P. vivax under the control of 5' flanking region of Pfdhfr and 3' UTR of Pfhrp2 ( Figure 1A). The 5' flanking region and truncated 5' coding sequence of Pfdhfr-ts are the sites for homologous recombination and replacement of Pfdhfr-ts with Pvdhfr-ts gene in the P. falciparum parasite chromosome 4.
Plasmodium berghei transfection plasmids were constructed from plasmids pL0017 [24] and pL0035 [28], which were kindly provided by Chris Janse and Andy Waters (Leiden University Medical Center, the Netherlands). In order to generate a P. berghei parasite line stably expressing wild-type PvDHFR-TS enzyme, transfection plasmids were modified to consist of two expression cassettes ( Figure 1B). The first cassette has the wild-type Pvdhfr-ts gene flanked by 5' and 3'UTR Pbdhfr-ts gene sequences. The second cassette is a drug selection cassette containing a fusion gene of positive (human dihydrofolate reductase; hdhfr) and negative (yeast yfcu) selectable markers under the control of 5' flanking region of P. berghei elongation factor 1a-a (Pbef1a-a) and 3'UTR sequence of Pbdhfr-ts gene.
For generation of P. berghei parasite stably expressing double mutant (S58R+S117N; SP21) PvDHFR-TS enzyme, the transfection plasmid contains the double mutant Pvdhfr-ts gene (Pvsp21) flanked by the Pbdhfr-ts gene 5' and 3' flanking sequences ( Figure 1C). The Pbdhfr-ts gene sequences serve as the sites for doublecrossover homologous recombination. The pyrimethamine-resistant Pvsp21 gene itself serves as a positive drug selectable marker for integration and gene replacement.

Generation of transgenic parasite lines stably expressing PvDHFR-TS Plasmodium falciparum transfection
Transfection of P. falciparum parasite was performed following a standard protocol [29]. Briefly, synchronized ring stage P. falciparum parasite strain K1CB1 (pyrimethamine-resistant) was transfected with 100 µg of transfection plasmid by electroporation (0.2 cm-gap electroporation cuvette, 0.310 kV, 950 µF). The transgenic parasites were first selected with blasticidin. Once blasticidin-resistant parasites were obtained, they were further treated with the negative selectable drug 5-fluorocytosine (5-FC) to select for double crossover integration of Pvdhfr-ts gene replacing the endogenous Pfdhfrts gene. After several rounds of drug cycling, stable transgenic parasites with the desired integration event were obtained and cloned by limiting dilution.

Plasmodium berghei transfection
Transfection of P. berghei parasites was performed using an Amaxa Nucleofector device (Amaxa), following the standard protocol [30]. Purified schizonts of GFPexpressing P. berghei ANKA parasite line MRA-867 (PbGFP) were transfected with 10 µg of linearized transfection plasmid in 100 µl of Human T Cell Nucleofector solution (Amaxa) using programme U-33, and then injected intravenously into naïve BALB/c recipient mice. Twenty-four hours post-injection, blood smears were made from the host animals and the presence of parasites checked by Giemsa staining. Drug-resistant parasites were selected by pyrimethamine administered to the drinking water (70 µg/ml). Under this selection regimen, PvDHFR-TS SP21 conferred pyrimethamine resistance to transgenic parasites. For selection of P. berghei harbouring wild-type Pvdhfr-ts, after pyrimethamine selection for the positive drug marker hdhfr, infected mice were injected intra-peritoneally twice a day with 5-FC (dissolved in 0.9% NaCl) at 1 g/kg bodyweight for a period of three days to select for parasites in which the drug selectable markers had been excised by single homologous recombination via the Pbdhfr-ts 3'UTR repeated sequence elements [28], whilst retaining the wild-type Pvdhfr-ts gene. The PvDHFR-TS-expressing transgenic P. berghei parasites were subsequently cloned by the method of limiting dilution [31].

Characterization of transgenic parasites
Correct integration of the constructs into the host's endogenous dhfr-ts locus was demonstrated by Southern analysis of restricted genomic DNA from transgenic parasites. Genomic DNA from erythrocytes infected with transgenic P. falciparum harbouring Pvdhfr-ts gene was purified using a QIAamp Blood Mini Kit (Qiagen) according to the manufacturer's instructions. Blood from mice infected with transgenic P. berghei harbouring Pvdhfr-ts or Pvsp21 genes was collected by cardiac puncture, and then depleted of leucocytes by passing through a CF11 cellulose column (Whatman). Following lysis of the erythrocyte pellet with 0.2% (w/v) saponin in PBS, total genomic DNA from parasites was obtained using a QIAamp Blood Mini Kit (Qiagen). The resulting DNA was then used for Southern analysis to confirm allelic replacement of the endogenous dhfr-ts with the P. vivax homologue. Briefly, 30 µg of genomic DNA from transgenic parasites was restriction-digested, separated in 0.6% (w/v) agarose gel and transferred onto a Hybond Figure 1 Diagrammatic representation of strategy to generate transgenic parasites by homologous recombination of Pvdhfr-ts gene replacing endogenous dhfr-ts locus. (A) Generation of transgenic P. falciparum stably expressing wild-type PvDHFR-TS. The endogenous Pfdhfrts locus was targeted via double homologous recombination with a plasmid containing expressing cassettes of wild-type Pvdhfr-ts, and blasticidin-S deaminase (bsd) marker. Yeast cytosine deaminase-uracil phosphoribosyl transferase (yfcu) was used as negative selectable marker. (B) Generation of transgenic P. berghei stably expressing wild-type PvDHFR-TS. The endogenous Pbdhfr-ts locus was targeted with a linearized plasmid containing expressing cassettes of wild-type Pvdhfr-ts, and fusion gene of human dhfr (hdhfr) and yeast yfcu as positive and negative selectable markers respectively. After double homologous recombination, the drug selectable markers are excised by single homologous recombination via the Pbdhfr-ts 3'UTR repeated sequence elements (blue boxes), whilst retaining the wild-type Pvdhfr-ts gene. (C), Generation of transgenic P. berghei stably expressing mutant PvDHFR-TS SP21. The endogenous Pbdhfr-ts locus was targeted with a linearized plasmid containing expressing cassettes of mutant Pvdhfr-ts sp21 (Pvsp21) flanked with 5' and 3' UTR of Pbdhfr-ts which also serve as the sites for double homologous recombination. Specific probes for Southern analysis are located by bold horizontal line. E: EcoRI, H: HindIII, K: KpnI. N+ nylon membrane (GE Healthcare) by capillary blotting. Digested DNA was fixed to the membrane by UV cross-linking and hybridized with specific probes prepared with a DIG labelling kit (Roche Applied Science). All probes used in this study are shown schematically in Figure 1. The pattern of hybridization was detected using alkaline phosphatase-conjugated anti-DIG antibody and CSPD reagent according the manufacturer's instructions (Roche Applied Science).
Validation of transgenic parasites as models for PvDHFR-TS screening SYBR Green I assay Plasmodium falciparum strains TM4/8.2, K1CB1 and transgenic P. falciparum expressing PvDHFR-TS were maintained in human erythrocytes as previously described and synchronized to 1% ring stages at 2% haematocrit. Ninety microliters of cultured parasites were transferred to each well of a 96-well microtitre plate before 10 μl of pyrimethamine (Sigma), chloroquine (Sigma) and dihydroartemisinin (a gift from Dafra Pharma International) at different concentrations were added in triplicate. The mixtures were incubated at 37°C under 5% CO 2 , 1% O 2 and 94% N 2 for 48 hours. To determine parasite growth, 100 μl of SYBR green I (at 1:5,000 dilution in culture media; Invitrogen) was added to the culture mixtures for 30 minutes at room temperature in the dark [32]. The SYBR green I-stained parasitized erythrocytes were then analysed using a Cytomics 500 FC MPL flow cytometer (Beckman Coulter) equipped with a 96-well plate adaptor. Erythrocytes were identified on the basis of their specific forward (FSC) and side (SSC) light-scattering properties. The green fluorescence signal was excited with an argon ion laser at a wavelength of 488 nm and the emission of the green fluorescence was detected using a 530/30 nm band pass filter. A previous analysis of non-infected erythrocytes was performed to determine a cut off for these cells to distinguish them from parasitized cells. The fluorescent intensity of a total of 100,000 events per sample was measured and data analysis was subsequently performed using CXP software (Beckman Coulter). For calculation of the percentage of parasite growth, the mean parasitaemia of untreated parasites was set at 100% growth. The concentration of drugs that inhibits 50% of the parasite growth (IC 50 ) was determined from the sigmoidal curve obtained by plotting the percentages of parasite growth against drug concentrations.

GFP assay
In order to validate transgenic P. berghei parasite stably expressing PvDHFR-TS enzymes as in vivo models for antifolate drug screening, the assessment of anti-malarial drug efficacy in vivo was performed using the standard four-day suppressive test [33]. Groups of at least three mice were inoculated with approximately 1 × 10 7 parasitized erythrocytes intravenously and then treated with standard anti-malarials: pyrimethamine (Sigma), chloroquine (Sigma) or artesunate (Mekophar Chemical Pharmaceutical) once daily for four successive days from the day of parasite inoculation. The drugs at chosen doses (mg/kg of body weight) were freshly prepared in 100% DMSO (pyrimethamine and artesunate) or distilled water (chloroquine) and administered orally by gavage. It was found that pyrimethamine at 50 mg/kg was lethal to mice as all animals died soon after treatment with this dose by oral administration (data not shown). Hence, the maximum pyrimethamine dose was 30 mg/ kg. Untreated controls were given either distilled water or DMSO only. The course of parasitaemia in untreated mice (control group) and in mice treated with different doses of drugs was monitored by flow cytometry since these parasites express GFP as fluorescence marker. A drop of blood was collected directly into wells of a 96well microtitre plate containing 200 µl of RPMI-1640 culture media and heparin (1U/µl). The GFP signal in these samples was measured as described above. For calculation of the growth inhibitory curves, the mean parasitaemia of the control samples was set at 0% inhibition. The concentration of drugs that inhibits 50% of the parasite growth (ED 50 ) was determined from the sigmoidal curve obtained by plotting the percent parasite growth inhibition against drug concentrations.

Statistical analysis
Parasites' growth and growth inhibitory curves and statistical analysis of the data were performed using the GraphPad Prism software version 5.0 (GraphPad Software). The non-linear regression function for sigmoidal dose-response (variable slope) was used to calculate the best-fit 50% inhibition concentration (IC 50 ) or 50% effective concentration (EC 50 ) values.

Generation of transgenic Plasmodium falciparum harbouring Pvdhfr-ts gene
Pyrimethamine-resistant P. falciparum strain K1CB1 was transfected with plasmid DNA for replacement of the endogenous Pfdhfr-ts with wild-type Pvdhfr-ts. Following rounds of positive and negative selections, transgenic P. falciparum were obtained. A clonal parasite line designated PfPvDTclB2 was obtained and gene replacement verified by Southern blotting analysis. As shown in Figure 2A, hybridization with a Pvdhfr-ts DNA probe showed the expected 3.7 kb and 5.9 kb EcoRI-digested fragments from transgenic PfPvDTclB2 genomic DNA and control plasmid DNA, respectively, confirming the integration of Pvdhfr-ts at the correct site, while no signal was detected for parental P. falciparum K1CB1 line. Another DNA probe specific to Pfdhfr could detect the expected 8.8 kb, 8.0 kb and 5.4 kb EcoRI-digested fragments from genomic DNA of transgenic PfPvDTclB2, parental P. falciparum K1CB1 line and control plasmid DNA, respectively. From these results, it can be concluded that transgenic PfPvDTclB2 parasites stably express wild-type Pvdhfr-ts instead of endogenous Pfdhfr-ts; furthermore, these parasites are clonal, lack episomal plasmid DNA and can be stably maintained without drug selection.

Generation of transgenic Plasmodium berghei harbouring Pvdhfr-ts gene
Blood stages of the reference GFP-expressing P. berghei parasite line PbGFP were transfected with the linear form of the transfection vectors by electroporation in order to introduce either wild-type Pvdhfr-ts or mutant Pvsp21 genes replacing the endogenous Pbdhfr-ts gene. For selection of P. berghei harbouring wild-type Pvdhfrts, a two-step drug selection procedure was performed. First, transgenic parasites with integrated wild-type Pvdhfr-ts were selected with pyrimethamine. Subsequently, negative selection was performed to select for parasites which had excised the positive-negative drug selection cassette in order to obtain parasites that have only wild-type Pvdhfr-ts gene integrated to the P. berghei genome ( Figure 1B). Transgenic P. berghei parasite clones stably expressing wild-type PvDHFR-TS (designated PbPvDTcl4) or double mutant PvDHFR-TS SP21 (designated PbPvSP21cl2) were obtained and analysed by Southern blotting. As shown in Figure 2B, the KpnI/ HindIII digested genomic DNA fragments corresponding to the parental wild type and transgenic P. berghei parasites (1.5 kb and 3.3 kb, respectively) were detected by a DNA probe specific to 3'UTR Pbdhfr-ts. Thus, this confirmed the successful generation of transgenic P. berghei parasites stably harbouring wild-type Pvdhfr-ts or mutant Pvsp21, replacing the endogenous Pbdhfr-ts gene.
Drug sensitivity analysis of transgenic Plasmodium falciparum stably expressing wild-type PvDHFR-TS enzyme Transgenic PfPvDTclB2 parasite was evaluated against standard anti-malarial drugs. Pyrimethamine is the standard antifolate drug and used as the primary compound to validate this system. As shown in Figure 3A and Table 1, the transgenic PfPvDTclB2 parasite was much more sensitive to pyrimethamine than the parental K1CB1 line, verifying that the wild-type Pvdhfr-ts gene replacing the Pfdhfr-ts gene is a pyrimethamine-sensitive variant. Moreover, the level of pyrimethamine sensitivity in the transgenic PfPvDTclB2 parasite is the same as the antifolate-sensitive P. falciparum TM4/8.2 strain (IC 50 = 0.03 ± 0.02 µM). The parental K1CB1 line is also resistant to chloroquine, a 4-aminoquinoline drug that inhibits haemozoin formation in the food vacuole of the parasites. This transgenic PfPvDTclB2 line shows the same chloroquine-resistant phenotype as the parental P. falciparum K1CB1 strain, with IC 50 values of 49.5 ± 5.8 nM and 46.0 ± 3.1 nM respectively ( Figure 3B and Table 1), indicating that the dhfr-ts gene replacement did not affect sensitivity to drugs not targeting DHFR-TS. Another non-antifolate drug control used in this study was dihydroartemisinin (DHA). All parasites tested in this study were sensitive to DHA at the IC 50 values of 0.6 ± 0.1 nM, 0.7 ± 0.3 nM and 0.4 ± 0.1 nM for P. falciparum TM4/8.2, K1CB1 and transgenic PfPvDTclB2, respectively ( Figure 3C and Table 1). Drug sensitivity analysis of transgenic Plasmodium berghei stably expressing PvDHFR-TS enzymes After inoculation, mice in the untreated control group showed a progressively increasing parasitaemia, and all the mice died by day 11 (data not shown). As shown in Figure 4A and Table 1, the transgenic PbPvDTcl4 demonstrated a drug susceptibility profile similar to that of the wild-type parental PbGFP with an ED 50 of 0.53 ± 0.24 mg/kg and 0.69 ± 0.21 mg/kg, respectively. This demonstrated that the wild-type PvDHFR-TS enzyme was equally susceptible to the antifolate compound compared with wild-type PbDHFR-TS. In contrast, transgenic PbPvSP21cl2 was approximately 40-fold more resistant to pyrimethamine than the PbPvDTcl4 parasite line ( Figure 4A). Therefore, the double mutant P. vivax DHFR-TS confers a high level of resistance to pyrimethamine in P. berghei.
The transgenic P. berghei lines were also tested with the non-antifolate drugs chloroquine and artesunate. As shown in Figure 4B and Table 1, all parasite lines were similarly susceptible to chloroquine. The ED 50 values against chloroquine were 1.56 ± 0.12 mg/kg, 2.85 ± 0.17 mg/kg, and 3.88 ± 0.13 mg/kg mg/kg in parental PbGFP, PbPvDTcl4 and PbPvSP21cl2 parasites, respectively. Artesunate is extremely potent against pyrimethamine-resistant parasites. It is a semi-synthetic derivative of artemisinin that is water-soluble and may therefore be given by injection. All parasite lines were also susceptible to artesunate treatment with ED 50 values of 5.43 ± 0.42 mg/kg, 7.43 ± 0.30 mg/kg and 7.59 ± 0.33 mg/kg in parental PbGFP, PbPvDTcl4 and PbPvSP21cl2 parasites, respectively (Figure and Table 1).

Discussion and Conclusions
This study describes the generation of both in vitro and in vivo transgenic Plasmodium parasites stably expressing P. vivax DHFR-TS enzyme and the application of these parasites for assessing antifolate compound efficacy. The transgenic Plasmodium parasites harbour only one copy of Pvdhfr-ts gene, which is under the control of the endogenous dhfr-ts regulatory sequences. The fact that it is possible to replace the endogenous Plasmodium dhfr-ts with a homologue from a different species suggests that dhfr-ts function is conserved among species of this genus. This is in agreement with biochemical evidence that cross-species heterodimers of Plasmodium DHFR-TS are fully functional [34]. While the expression of double mutant PvDHFR-TS SP21 was not deleterious and can support the growth of the transgenic P. berghei parasites, attempts to obtain P. falciparum stably expressing double mutant PvDHFR-TS SP21 were not successful (data not shown). The reasons may be the catalytic activity of PvDHFR-TS SP21 is insufficient to support the growth of transgenic P. falciparum. It was reported that the catalytic activity, as determined by k cat /K m of H2folate , of PvDHFR SP21 is approximately 13 times and 27 times lower than that of wild-type PvDHFR [11] and wild-type PfDHFR [35], respectively.
A major advantage of the Pvdhfr-ts transgenic parasites generated in the study is that the gene is stably maintained in an integrated fashion. This issue is important since it has been reported that the apparent copy number of plasmid DNA in episomally transfected Plasmodium is variable, which may affect the drug sensitivity profile in drug testing assays. The reasons for varying copy numbers are related to the nature of the episomal plasmid itself and the level and duration of drug pressure [36]; moreover, higher drug concentration are known to select for increased copy numbers of the episome [37]. In the absence of drug pressure, both transgenic P. falciparum and P. berghei lose episomal transgenes [36,38,39]; hence, constant drug pressure is needed to maintain the episomes. In the earlier report of P. falciparum episomally expressing Pvdhfr-ts, no effect of episomal copy number on the drug sensitivity profile was observed, probably owing to the tight regulation of the parasite DHFR-TS expression level [22,40]. However, the episomal copy number did vary since the drug used to maintain the episome must be withdrawn for compound testing, hence the episomal system may not be sufficiently robust as a general drug screening method.
In this study, in vitro and in vivo Plasmodium models have been generated specifically for anti-PvDHFR screening. There are reports that compounds that have anti-P. falciparum activity in in vitro screening do not show the same anti-Plasmodial activity in in vivo models, such as P. berghei [41,42]. It is therefore premature to assume that compounds found to be active in vitro will also be efficacious in vivo before pharmacokinetic studies of the compounds in animal models have been conducted, since compounds with poor pharmacokinetic properties may simply not reach their targets. At present, the only in vivo model for P. vivax is P. cynomolgi infecting primates [43]. This model is expensive and  restricted to specialized laboratories; hence the transgenic P. berghei expressing P. vivax targets is an attractive alternative experimental model. In conclusion, transgenic P. falciparum and P. berghei stably expressing PvDHFR-TS replacing the endogenous Plasmodium dhfr-ts were successfully generated as in vitro and in vivo Plasmodium drug screening models. The drug sensitivities of the transgenic lines varied according to the introduced Pvdhfr-ts gene. Until a routine continuous in vitro culture system is developed for P. vivax, these transgenic models will be useful for screening of anti-P. vivax compounds targeting the PvDHFR enzyme. A similar approach could be used to generate transgenic models specific for other targets of interest, thus facilitating the development of anti-P. vivax drugs in general.