A high-throughput screen targeting malaria transmission stages opens new avenues for drug development

Abstract

A major goal of the worldwide malaria eradication program is the reduction and eventual elimination of malaria transmission. All currently available antimalarial compounds were discovered on the basis of their activity against the asexually reproducing red blood cell stages of the parasite, which are responsible for the morbidity and mortality of human malaria. Resistance against these compounds is widespread, and there is an urgent need for novel approaches to reduce the emergence of resistance to new antimalarials as they are introduced. We have established and validated the first high-throughput assay targeting the red blood cell parasite stage required for transmission, the sexually reproducing gametocyte. This assay will permit identification of compounds specifically targeting the transmission stages in addition to the asexual stage parasites. Such stage-specific compounds may be used in a combination therapy, reducing the emergence of resistance by blocking transmission of resistant parasites that may be selected in a patient.

Figure 1.

Characterization of transgenic parasite lines expressing GFP under different gametocyte-specific promoters. A—C, GFP expression during sexual development. Transgenic parasites were cultured in T75 flasks, induced for in vitro sexual development by reduction of hematocrit as described elsewhere [13], and analyzed live by fluorescence microscopy. Morphological differentiation of individual gametocyte stages (stage I–V) was based on the classification by Hawking et al [43]. GFP was expressed throughout sexual development starting at sexually committed schizont stages. In 744/GFP (A) and 748/GFP (B), fluorescence peaks during early development. In 164/GFP (C), fluorescence intensity increases during development, either through elevated expression levels and/or accumulation of GFP over time. Stage I gametocytes are indistinguishable from asexual trophozoites except for the presence of dispersed hemozoin crystals in the former. Stage II gametocytes adopt a characteristic morphology, with an elongated wheat-shaped body and subpellicular extensions. SI/SII, gametocyte stage I and II. D, Co-localization of GFP fluorescence in 164/GFP with a transmission stage marker. To determine whether GFP expression was indeed gametocyte specific, we performed immunofluorescence assay in induced cultures using antibodies against the sexual stage antigen Pfs16 for co-localization. Two representative panels are shown (D), with a stage I gametocyte (top) and a stage III gametocyte (bottom).

Figure 2.

Assay protocol in 96-well plates. Parasites were first treated with sorbitol to obtain a loosely synchronous ring stage population. After 1 cycle, sequential purification of late stages on a Percoll gradient and treatment with sorbitol 3 h later, after reinvasion, resulted in a highly synchronous population of ring-stage parasites. These were seeded at 1% parasitemia and 4% hematocrit into a 96-well plate. After 24 h, the hematocrit was reduced to 2% (ie, sexual conversion induced) by doubling the volume of culture medium from 110 μL to 220 μL. After another 72 h and 1 asexual replication cycle, cells from each well were analyzed to quantify total parasite and gametocyte loads. The timeline represents the assay period (96 h) performed in 96-well plates.

Figure 3.

Establishment of Hoechst and GFP detection by cytometry. Upper panel, Hoechst staining of infected red blood cells (iRBCs) at day 2 of the assay demonstrates that the majority of parasites have 1 nucleus (ie, rings and gametocytes [R + G]), whereas a subset is multinucleated (trophozoites and schizonts [T + S]). Both fresh uninfected red blood cells (uRBCs; from 4°C) and uRBCs incubated along for the assay period and stained with Hoechst were used as a negative control for gating. Lower panel, iRBCs at day 2 of the assay show GFP–positive population. This population is significantly higher than the background signal represented by autofluorescence in uRBC control samples that were either fresh or were incubated for the entire assay period.

Figure 4.

Definition of assay parameters. A, Signal-to-noise ratios using uninfected red blood cells (uRBCs) and the parental P2G12 line as potential negative controls. Signal-to-noise ratios were calculated described in Materials and Methods. The ratio drops from 30-fold (when using cells from day -1 as negative control) to 5-fold (when using cells from day 2). Each data point represents at least 3 biological replicates done with technical duplicates. B, Comparison of microscopy and flow cytometry (FCM) for the quantification of parasite and gametocyte loads. An increase in parasite concentration from initially 1% to ∼18% during the assay period of 96 h is detected both by counting Giemsa-stained blood smears (circles, black solid line) and by quantifying a Hoechst-stained sample of the same culture by FCM (cubes, black dashed line). Sexual stages, defined as the GFP-positive subpopulation of infected red blood cells, increased from background levels of ∼0.005% to ∼0.15% at day 2, representing a 30-fold absolute or 5-fold relative increase (correcting for replication rate). FCM detects only GFP expression of stage II gametocytes (cubes, green dashed line), whereas the weaker fluorescence of earlier stages (ie, the sexually committed ring and stage I; termed Pre-Stage II in the graph) is only detectable by microscopy (triangle, solid blue). Stage II gametocyte numbers quantified by microscopy (circles, solid green line) correlate well with GFP-positive cells determined by FCM. Each data point represents at least 3 biological replicates done with technical triplicates. The graphs show mean values of the experiments, whereas the error bars represent the standard error of the mean. C, Sensitivity of microscopy and FCM to quantify parasite load (left panel) and gametocyte load (right panel). Each data point was prepared in duplicates as described in Materials and Methods and analyzed in parallel by FCM (GFP and Hoechst) as well as fluorescence microscopy (GFP and Giemsa). The resulting measured parasite and gametocyte loads were plotted against the calculated values of the serial dilution. The corresponding R² values are as follows: 0.9659 by microscopy and 0.9798 by FCM for parasite load; 0.8345 by microscopy and 0.9610 by FCM for gametocyte load. D, The transgenic parasite line 164/GFP shows no loss of sexual conversion during continuous culture. The box-and-whisker plot represents all assay positive control wells over the time period when drug assays were performed. The sample sizes are as follows: month 0.5, n = 16; month 1, n = 18; month 3, n = 36; month 3.5, n = 22; month 4, n = 32; month 4.5, n = 39; month 5, n = 29; month 5.5, n = 45; month 6, n = 48; and month 6.5, n = 24. No experiments were performed in months 1.5–2.5.

Figure 5.

Half maximal inhibitory concentration curves representing the effect of chloroquine (A), DHA (B), atovaquone (C), and methylene dlue (D) on gametocyte and parasite loads. Each assay represents at least 3 biological replicates with 2 technical replicates per plate. The graphs show mean values of the experiments, whereas the error bars represent the standard error of the mean. conc., concentration.