P. falciparum in vitro killing rates allow to discriminate between different antimalarial mode-of-action

Abstract

Chemotherapy is still the cornerstone for malaria control. Developing drugs against Plasmodium parasites and monitoring their efficacy requires methods to accurately determine the parasite killing rate in response to treatment. Commonly used techniques essentially measure metabolic activity as a proxy for parasite viability. However, these approaches are susceptible to artefacts, as viability and metabolism are two parameters that are coupled during the parasite life cycle but can be differentially affected in response to drug actions. Moreover, traditional techniques do not allow to measure the speed-of-action of compounds on parasite viability, which is an essential efficacy determinant. We present here a comprehensive methodology to measure in vitro the direct effect of antimalarial compounds over the parasite viability, which is based on limiting serial dilution of treated parasites and re-growth monitoring. This methodology allows to precisely determine the killing rate of antimalarial compounds, which can be quantified by the parasite reduction ratio and parasite clearance time, which are key mode-of-action parameters. Importantly, we demonstrate that this technique readily permits to determine compound killing activities that might be otherwise missed by traditional, metabolism-based techniques. The analysis of a large set of antimalarial drugs reveals that this viability-based assay allows to discriminate compounds based on their antimalarial mode-of-action. This approach has been adapted to perform medium throughput screening, facilitating the identification of fast-acting antimalarial compounds, which are crucially needed for the control and possibly the eradication of malaria.

Figure 1. Schematic representation of the in vitro PRR assay.

A. Intraerythrocytic P. falciparum cultured at 0.5% parasitemia and 2% hematocrit is treated with drugs. The medium is exchanged and the drug replenished every 24 hours. Aliquots corresponding to 10⁵ parasites are taken out at defined time points, washed, and free-drug parasites cultured with fresh erythrocytes under limiting serial dilution conditions (see Material and Methods). Parasite growth is subsequently monitored after 21 days and confirmed after 28 days, allowing to calculate the initial number of viable parasite in the aliquot. B. Parasite viability measurement allows in turn to determine the drug lag phase (i.e. time needed to reach the maximal rate of killing), PRR over one life cycle, and 99.9% PCT (i.e. the time needed to decrease the number of viable parasites by 3 –log units). The data presented in this panel are for illustration purpose only. Axe Y shows log (viable parasites +1) to allow representation of logarithms when counting of number of viable parasites is equal to zero.

Figure 2. Comparison of metabolic and viability assays.

A. P. falciparum radio-labeled hypoxanthine incorporations after 48 hours of drug treatment at concentrations corresponding to 10× IC₅₀ with artemisinin, atovaquone, and azithromycin, reported as percentages of untreated controls. Data are averages of 12 repetitions from two independent experiments B. P. falciparum viability after 48 hours of drug treatment at concentrations corresponding to 10× IC₅₀ with artemisinin, atovaquone, and azithromycin, reported as number of viable parasite, as determined by limiting serial dilutions. Data are averages of 4 independent experiments. In both panels, error bars represent the standard error of the mean (SEM).

Figure 3. Parasite viability in response to various classical antimalarial drugs.

A. P. falciparum viability time-course profiles for chloroquine (chq), mefloquine (mef), piperaquine (pip), artemisinin (art), lumefantrine (lum), pyronaridine (pyro), pyrimethamine (pyri), and atovaquone (ato). Error bars represent the SEM of at least 4 independent experiments. B. and C. Scatter plots of the compounds tested reporting the IC₅₀ versus the log(PRR) and 99.9% PCT, respectively. The dotted line in panel B is a log linear regression, the slope thereof is not significantly different from zero (p = 0.48). Data of panel C do not converge enough to establish a regression line.

Figure 4. Parasite viability time-course in response to various concentrations of drugs.

A. P. falciparum viability time-course profiles for atovaquone, pyrimethamine, and artemisinin at concentrations corresponding to 1×, 3×, 10×, and 100× their respective IC₅₀. Error bars are SEM of at least 4 independent experiments. B. Values represented in panel A. No PRR or 99.9% PCT could be calculated for the 1× IC₅₀ conditions.

Figure 5. Classical antimalarial killing rate profiles.

P. falciparum viability time-course profiles of four classical antimalarial drugs, illustrating the wide range of speed-of-action measured. Viable parasite are cleared in response to fast-acting (art and chq), but not slow-acting (pyri and ato) drugs after 72 hours of treatment.

Figure 6. Parasite viability correlates with the mode-of-action of antimalarials.

P. falciparum viability time-course profiles for artemisinin, atovaquone, GW648495, and GW844520 measured at 0, 24, 48, 72, 96, and 120 hours. Artemether and artesunate have been investigated at 24 hours only. Error bars are SEM of at least 4 independent experiments.

Figure 7. Single time-point viability measurement.

P. falciparum viability after 72 hours of treatment with the indicated drugs, reported as the log of viable parasite+1, as compared to the untreated controls. The red line represents the threshold of 99.9% parasite reduction. The data are representative of at least 3 independent experiments.