Na+ Influx Induced by New Antimalarials Causes Rapid Alterations in the Cholesterol Content and Morphology of Plasmodium falciparum

Abstract

Among the several new antimalarials discovered over the past decade are at least three clinical candidate drugs, each with a distinct chemical structure, that disrupt Na+ homeostasis resulting in a rapid increase in intracellular Na+ concentration ([Na+]i) within the erythrocytic stages of Plasmodium falciparum. At present, events triggered by Na+ influx that result in parasite demise are not well-understood. Here we report effects of two such drugs, a pyrazoleamide and a spiroindolone, on intraerythrocytic P. falciparum. Within minutes following the exposure to these drugs, the trophozoite stage parasite, which normally contains little cholesterol, was made permeant by cholesterol-dependent detergents, suggesting it acquired a substantial amount of the lipid. Consistently, the merozoite surface protein 1 and 2 (MSP1 and MSP2), glycosylphosphotidylinositol (GPI)-anchored proteins normally uniformly distributed in the parasite plasma membrane, coalesced into clusters. These alterations were not observed following drug treatment of P. falciparum parasites adapted to grow in a low [Na+] growth medium. Both cholesterol acquisition and MSP1 coalescence were reversible upon the removal of the drugs, implicating an active process of cholesterol exclusion from trophozoites that we hypothesize is inhibited by high [Na+]i. Electron microscopy of drug-treated trophozoites revealed substantial morphological changes normally seen at the later schizont stage including the appearance of partial inner membrane complexes, dense organelles that resemble "rhoptries" and apparent nuclear division. Together these results suggest that [Na+]i disruptor drugs by altering levels of cholesterol in the parasite, dysregulate trophozoite to schizont development and cause parasite demise.

Fig 1. Structures of [Na⁺]_i disruptor compounds and their effect on protein synthesis.

(A) Structures of compounds discussed in the text and their EC₅₀ values. (B) Pyrazoleamide and spiroindolones do not inhibit parasite protein synthesis. 32–34 h PMI (post merozoite invasion) trophozoite stage P. falciparum 3D7 parasites were exposed to the vehicle (Control) or to 10x EC₅₀ (10 nM) of PA21A050, KAE609, cycloheximide (2000 nM), or artemisinin (100 nM) for 2h in a ³⁵S-methionine/³⁵S-cysteine containing medium. Autoradiograph of labeled total proteins in saponin-freed (B) and anthrolysin O (ALO)-freed parasites (C) are shown. Isolated parasites with saponin led to detection of a highly reduced amount of proteins in PA21A050 and KAE609 treated parasites but not when parasites were freed by ALO.

Fig 2. Pyrazoleamide and spiroindolone antimalarials induce saponin sensitivity to the PPM.

Trophozoite stage P. falciparum 3D7 (30–34 h post-infection) were exposed to the vehicle (Ctrl) or the indicated doses of PA21A050 (A) or KAE609 (B) for 2 h, followed by mild saponin treatment to release the parasites and subjected to Western blot analysis using antibodies to aldolase or Exp2. (C) Densitometric measures (using Image J) of the band intensities from the Western blots were plotted as functions of compound concentrations to derive EC₅₀ values for aldolase loss in a 2 h exposure. Growth inhibition by PA21A050 and KAE609 was assessed by ³H-hypoxanthine incorporation by P. falciparum in a 48 h assay. Saponin-freed parasites were exposed to 10 nM PA21A050 (D) or KAE609 (E) for the indicated time. Parasite proteins were subjected to SDS-PAGE without further treatment (upper panels in D and E) or after subjecting the parasites to a second saponin treatment (middle and lower panels in D and E). Western blots were probed with antibodies to aldolase or Exp2. Pyrazoleamide resistant line, Dd2-R21 parasitized erythrocytes were exposed to 10 nM PA21A050 (F) and 10 nM KAE609 (G) for the indicated time followed by saponin treatment and Western blotting and probing with antibodies to aldolase or Exp2. Error bars represent the standard deviation (SD) of the measurements. All western blot images are the representative of multiple biological replicates.

Fig 3. Pyrazoleamide and spiroindolone induced saponin sensitivity is due to cholesterol incorporation into the parasite.

(A and B) Cholesterol extraction with MβCD eliminates saponin sensitivity induced by the drugs, whereas cholesterol-loaded MβCD does not. Freed 30–34 h post-infection trophozoites were treated with either PA21A050 (A) or KAE609 (B). Parasites were treated with the drugs (lanes 1 and 2), or were exposed to MβCD to extract cholesterol and then treated with the drugs (lanes 3 and 4). Freed parasites were exposed to MβCD loaded with cholesterol followed by the drug treatment (lanes 5). Freed parasites were first treated with the drugs and then exposed to either MβCD (lanes 6 and 7) or MβCD loaded with cholesterol (lanes 8). After the indicated time, parasites were subjected to a brief exposure to saponin, centrifuged and subjected to SDS-PAGE followed by immunobloting with antibodies to aldolase and Exp2. (C and D) Cholesterol loaded MβCD can donate cholesterol to cholesterol-depleted freed parasites, restoring drug-induced saponin sensitivity. Experiments with PA21A050 treatment (C) and KAE609 treatment (D) are shown. Inclusion of just 0.625 mM cholesterol-loaded MβCD in cholesterol-depleted parasites restored saponin sensitivity to parasites when treated with the drugs. However, in absence of the drugs inclusion of cholesterol-loaded MβCD did not impart saponin sensitivity, suggesting an active process of cholesterol exclusion from the parasite.

Fig 4. Na⁺ influx induction is required for cholesterol acquisition by the parasite in PA21A050 and KAE609 treated parasites.

Trophozoite stage P. falciparum parasites (30–34 h post-infection), adapted to grow in low [Na⁺] medium, were treated with the indicated concentration of the PA21A050 (A) or KAE609 (B). Cholesterol acquisition was assessed by sensitivity to saponin-mediated loss of aldolase, which was absent in parasites grown in low [Na⁺] medium but apparent in parasites grown in normal medium. (C) Saponin sensitivity was not induced by the compounds in parasites at the ring stage where the erythrocyte cytosol does not yet contain a high [Na⁺] and the parasites are not exposed to the high [Na⁺] of the medium. Western blots were probed with antibodies to aldolase and Exp2. All western blot images are representative of multiple biological replicates.

Fig 5. A monovalent ionophore, maduramicin, causes rapid Na⁺ influx and cholesterol incorporation into the trophozoite stage P. falciparum.

(A) Growth inhibition by maduramicin was assessed by ³H-hypoxanthine incorporation by P. falciparum in a 48 h assay, with EC₅₀ of 0.44 nM. (B) Maduramicin causes rapid influx of Na⁺ into P. falciparum. Ratiometric measurements of [Na⁺]_i were carried out as described in Materials and Methods. Addition of 3.6 nM maduramicin rapidly caused Na⁺ influx into the parasite. (C) The plateau levels of [Na⁺]_i at different concentrations of PA21A050 and maduramicin are shown, indicating a dynamic balance between Na⁺ influx and Na ⁺ efflux. (D) P. falciparum 3D7 trophozoites (30–34 h post-infection) were exposed to the vehicle (Ctrl) or the indicated doses of maduramicin for 2 h, followed by the assessment of saponin sensitivity as described in Fig 2. Leakage of cytosolic aldolase was assessed by SDS-PAGE and immunobloting. (E) P. falciparum 3D7 trophozoites were exposed to the vehicle (Ctrl) or 5 nM maduramicin for the indicated period of time. Treated parasites were released by mild saponin treatment and subjected to Western blot analysis using antibodies to aldolase or Exp2.

Fig 6. PA21A050 and KAE609 induced Cholesterol incorporation into the parasite is rapidly reversible.

After 2 h of treatment, removal of PA21A050 (A) and KAE609 (B) rapidly leads to restoration of insensitivity to saponin-mediated aldolase loss within 30–60 min. Western blots were probed with antibodies to aldolase and Exp2. (C) After the treatment with PA21A050 (10 nM) or KAE609 (10 nM) for 2 h followed by the removal of compounds, ³H-hypoxanthine incorporation over a 24 h was assessed as a measure of parasite viability. (D) Parasite growth over a 96 h period following the 2 h compound treatment was assessed by measuring parasitemia in Giemsa stained smears every 24 h.

Fig 7. Reversible clustering of a GPI-anchored protein in compound treated trophozoite stage parasites.

(A) Distribution of the GPI-anchored protein MSP1 was examined by immunofluorescence assays in 32–34 h PMI (post merozoite invasion) trophozoite stage P. falciparum 3D7 exposed for 2 h to the vehicle (control), PA21A050 (10 nM) or KAE609 (10 nM). (B) After 2 h of the removal of the compounds, parasites largely restored the distribution of MSP1 throughout the PPM. (C) The compound-induced MSP1 clustering was not observed in parasites adapted to grow in low [Na⁺] medium following compound treatments for 2 h. (D) Quantitation of parasites showing MSP1 clustering following treatment with the indicated compound for 2 h from 3 biological replicates (the total number of parasites (N) assessed is indicated above in each treatment condition). Error bars are the SD of the percentage of clustered MSP1 parasites determined under each experimental condition (N = 3). (E) Immunogold labeling for MSP1 on a representative P. falciparum control parasite (upper panel), in which the gold particles are distributed throughout the parasite plasma membrane (PPM); and on a representative PA21A050-treated parasite (lower panel), in which the gold particles are clustered in one patch. Cryo-sections were labeled with the anti-MSP1 antibody, and binding revealed with protein A-gold particles (10 nm). PPM, Parasite plasma membrane; PVM, Parasitophorous vacuole membrane; IMC, inner membrane complex. Scale bars are 100 nm. In each immunogold labeling experiment, 60–70 parasites were imaged, and Fig 6E shows representative images.

Fig 8. Ultrastructural representation of parasites exposed to compounds.

Transmission electron micrographs of 32–34 h PMI (post merozoite invasion) trophozoite stage P. falciparum 3D7 infected erythrocytes. (A) untreated parasites and (B) parasites treated for 2 h with 10 nM KAE609 (panels a, c, e, f, i) or 10 nM PA21A050 (panels b, d, g, h, j) showing progressive steps in parasite transformation towards a schizogony-like stage characterized by scission of the nucleus (n; arrows in a), furrow cleavage demarked by the IMC (arrows in panels e, f) and secretory organelle biogenesis as illustrated for rhoptries (r). Scale bars are 250 nm.

Fig 9. A 2 h treatment with PA21A050 or KAE609 did not show increased DNA content in P. falciparum.

32–34 h PMI (post merozoite invasion) trophozoite stage P. falciparum 3D7 parasites were exposed to the vehicle (Control) (A) or 10 nM PA21A050 (B) or 10 nM KAE609 (C) for 2 h, fixed and stained with SYBR Green to assess DNA content, as measured by fluorescence intensity detected using flow cytometry. Uninfected erythrocytes are gated out and pseudocolor density dot plots of the infected parasite populations are depicted.

Fig 10. A schematic model showing Na⁺ dependent events in normal and drug-treated parasites.

Green arrows and hammers indicate the proposed normal physiological process whereas red arrows and hammers indicate events triggered by treatment with [Na⁺]_i disruptor compounds. Under physiological conditions, a signaling event at the initiation of schizogony is hypothesized to inhibit Na⁺ pumping by the PPM-localized PfATP4, resulting in Na⁺ influx within the parasite cytoplasm. The increased [Na⁺]_i provides further signaling that leads to the inhibition of a putative cholesterol pump, resulting in increased accumulation of cholesterol in the PPM; the increased cholesterol, in turn, is required for the appropriate display of merozoite surface proteins. The increased [Na⁺]_i also constitutes a signal for further progression of normal schizogony processes such as nuclear division and formation of the inner membrane complex. Spiroindolones and pyrazoleamides prematurely usurp this process by either directly inhibiting PfATP4 or by mimicking a schizogony signal, either of which would result in influx of Na⁺ into the parasite cytoplasm. Premature inhibition of the putative cholesterol pump and induction of schizogony processes by high [Na⁺] would result in parasite death.