Decreasing pfmdr1 copy number in plasmodium falciparum malaria heightens susceptibility to mefloquine, lumefantrine, halofantrine, quinine, and artemisinin

Abstract

The global dissemination of drug-resistant Plasmodium falciparum is spurring intense efforts to implement artemisinin (ART)-based combination therapies for malaria, including mefloquine (MFQ)-artesunate and lumefantrine (LUM)-artemether. Clinical studies have identified an association between an increased risk of MFQ, MFQ-artesunate, and LUM-artemether treatment failures and pfmdr1 gene amplification. To directly address the contribution that pfmdr1 copy number makes to drug resistance, we genetically disrupted 1 of the 2 pfmdr1 copies in the drug-resistant FCB line, which resulted in reduced pfmdr1 mRNA and protein expression. These knockdown clones manifested a 3-fold decrease in MFQ IC(50) values, compared with that for the FCB line, verifying the role played by pfmdr1 expression levels in mediating resistance to MFQ. These clones also showed increased susceptibility to LUM, halofantrine, quinine, and ART. No change was observed for chloroquine. These results highlight the importance of pfmdr1 copy number in determining P. falciparum susceptibility to multiple agents currently being used to combat malaria caused by multidrug-resistant parasites.

Figure 1

pfmdr1 knockdown strategy and molecular characterization of clones. The transfection plasmid pcamBSD^KD/mdr contains a 1.6-kb pfmdr1 fragment (Δmdr ) centrally located in the coding sequence and a blasticidin-S-deaminase (bsd) selectable marker cassette (A). Single-site crossover between the plasmid and 1 of the 2 endogenous pfmdr1 copies results in inactivation of the copy, leaving 1 functional pfmdr1 copy remaining (B and C ). The disrupted locus contains an upstream pfmdr1 fragment lacking the 3′ end of the gene and the 3′ untranslated region (UTR) as well as a downstream pfmdr1 fragment lacking the 5′ UTR and the 5′ start of the gene, with these 2 fragments separated by the bsd selectable marker. Polymerase chain reaction (PCR) primers (P1–P4) and BamHI (B) fragment sizes are indicated. Square brackets delineate the plasmid sequence that can integrate as tandem linear copies (n ≥ 1). PCR analyses of the parental FCB line and the knockdown clones (KD1^mdr1, KD2^mdr1, and KD3^mdr1) with primers specific for either the upstream truncated locus (P1+P2) or the downstream remnant (P3+P4) confirmed the disruption of 1 copy of pfmdr1 (D and E ). Southern blot hybridization of BamHI-digested genomic DNA samples with a bsd probe revealed 12.5-kb and 6.1-kb bands in the recombinants only, which is consistent with integration of tandem plasmid copies into the pfmdr1 locus (F). The no. of plasmid copies that integrated in tandem was estimated by densitometry to be 3 for KD1^mdr1 and KD2^mdr1 and 2 for KD3^mdr1, indicating that the integration event that occurred in KD3^mdr1 was distinct from that which occurred in the 2 other clones. pBS, pBluescript.

Figure 2

Quantitative real-time reverse-transcription polymerase chain reaction (RT-PCR) analysis of the parental FCB line and the pfmdr1 knockdown clones. The effect that disruption of 1 copy of pfmdr1 had on transcript levels was determined by quantitative real-time PCR using cDNA prepared from tightly synchronized parasites. pfmdr1 transcript levels for each are presented as mean ± SE normalized genome equivalents [27]. Panel A shows an ~80% reduction in pfmdr1 transcript levels for the clones KD1^mdr1 and KD2^mdr1, compared with that for the FCB line, as determined from early ring-stage preparations (when transcription is maximal). Results for KD3^mdr1 were excluded, because of low real-time PCR yields. Panel B shows the results for parallel quantitative real-time RT-PCRs, which revealed no significant change in pfcrt transcript levels between the parental line and the knockdown clones.

Figure 3

Western blot analysis of the parental FCB line and the pfmdr1 knockdown clones. Probing of sorbitol-synchronized trophozoite-stage proteins with anti-PfMDR1 antibodies revealed an ~160-kDa band in both the parental line and the clones (A). Densitometric analysis revealed that PfMDR1 expression levels (normalized to PfERD2 expression levels) were reduced by 38%, 61%, and 52% for the KD1^mdr1, KD2^mdr1, and KD3^mdr1 clones, respectively. Densitometric analysis of these PfERD2 bands was indicative of essentially equivalent protein loadings across the parental FCB line and the knockdown clones, with the clones displaying signals that were 85%–106% of those of the FCB line (B).

Figure 4

In vitro antimalarial drug response of the pfmdr1 knockdown clones. In vitro [³H]-hypoxanthine incorporation assays (72 h) were performed with the knockdown clones and the parental FCB line, which were tested in duplicate against each antimalarial drug on 3–5 separate occasions. IC₅₀ values (shown as means ± SEs) were derived by regression analysis. Numerical values are listed in table 1. For statistical comparisons, 1-way analysis of variance tests with Bonferroni posttests were performed. *P < .05, **P < .01, and ***P < .001, compared with the FCB parental line.