Revealing the sequence and resulting cellular morphology of receptor-ligand interactions during Plasmodium falciparum invasion of erythrocytes

Abstract

During blood stage Plasmodium falciparum infection, merozoites invade uninfected erythrocytes via a complex, multistep process involving a series of distinct receptor-ligand binding events. Understanding each element in this process increases the potential to block the parasite's life cycle via drugs or vaccines. To investigate specific receptor-ligand interactions, they were systematically blocked using a combination of genetic deletion, enzymatic receptor cleavage and inhibition of binding via antibodies, peptides and small molecules, and the resulting temporal changes in invasion and morphological effects on erythrocytes were filmed using live cell imaging. Analysis of the videos have shown receptor-ligand interactions occur in the following sequence with the following cellular morphologies; 1) an early heparin-blockable interaction which weakly deforms the erythrocyte, 2) EBA and PfRh ligands which strongly deform the erythrocyte, a process dependant on the merozoite's actin-myosin motor, 3) a PfRh5-basigin binding step which results in a pore or opening between parasite and host through which it appears small molecules and possibly invasion components can flow and 4) an AMA1-RON2 interaction that mediates tight junction formation, which acts as an anchor point for internalization. In addition to enhancing general knowledge of apicomplexan biology, this work provides a rational basis to combine sequentially acting merozoite vaccine candidates in a single multi-receptor-blocking vaccine.

Fig 1. Merozoite invasion kinetics are generally similar for different P. falciparum strains.

(A) Cartoon of merozoite ligands, their erythrocyte receptors, and methods used to ablate these interactions. (B-G) Box and whisker plots; line indicates median, box indicates 75–25% range, and ‘whiskers’ denote total data range. Significant differences (p<0.05) are shown on the graphs, differences approaching significance (0.05>p<0.1), and insignificant p values are noted in figure legends. The number of (n) invasions (B-G) and contacts (B-D) is shown below each graph. (B) Percent of merozoite—erythrocyte (RBC) contacts resulting in successful invasion is lower in Δ175 vs. other strains used in this study except for W2m (W2mef vs. Δ175 p = 0.151). Comparisons between all other strains were not significant (p = 0.621–0.891). (C) Both the number of erythrocytes contacted before merozoite invasion (white) and total erythrocyte contacts (as some were contacted more than once (grey)), do vary between strains. This approaches significance when comparing Δ175 vs. D10 (p = 0.062) and W2m (p = 0.052) in number of erythrocytes contacted, and Δ175 vs. D10 (p = 0.071) in total number of interactions. All other comparisons p = 0.605–0.764. (D) The average number of invasions per schizont rupture is nearly the same for all strains tested, however approaches a significant difference when comparing D10 vs. 3D7 (p = 0.056) and Δ175 (p = 0.055). All other comparisons p = 0.208–0.918. (E) The time from first contact of merozoite and erythrocyte to start of internalisation is nearly the same for all strains tested except when comparing D10 vs. 3D7. All other comparisons p = 0.107–0.960. (F) The time from start to finish of internalisation is nearly the same for all strains tested. All comparisons p = 0.592–0.970. (G) The time from finish of internalisation to start of echinocytosis is very similar for all strains tested. All comparisons p = 0.421–0.800.

Fig 2. Strong merozoite deformation is associated with successful merozoite invasion.

(A) Representative images demonstrating levels of merozoite deformation of erythrocytes. (B) Diagram illustrating the derivation of deformation scores. (C) The percentages of each deformation score for the parasite strains used. The total scores are shown first followed by breakdown into invading and non-invading merozoites. The number (n) of interactions scored for each strain is shown below the columns. (D) There is no difference in deformation scores between the strains used, however in all strains the invading merozoites deformed their target erythrocytes significantly more than non-invading merozoites (last column).

Fig 3. Blocking specific receptor-ligand interactions can decrease deformation, invasion and echinocytosis of the host erythrocyte.

(A, B) Stacked graphs showing the number of invasions per rupture on top and matching deformation scores underneath. The parasite lines and treatments are shown at the bottom with untreated parasites in lanes 1–4 and parasites treated with invasion inhibitors in lanes 5–15. Note that data from 3C has been reproduced in lanes 1–4 of B in this figure, as controls for the invasion inhibitory treatments. (C) Table shows the parasite strain, the type of treatment, its target interaction, the number (n) of interactions counted, and whether echinocytosis (EC) of the contacted erythrocyte occurred. P values indicate whether the deformation scores for the treated parasites were significantly different (p<0.05) to untreated parasites. Note that Δ175 has a p value because it has been compared to the parental W2mef line.

Fig 4. The parasite’s actin-myosin motor is involved in rapid host cell selection.

(A) The number of merozoite contacts per schizont rupture which result in invasion in untreated parasites is the same as the number of merozoite contacts which result in echinocytosis in the invasion-blocked cytD treated parasites. (B) The percentage of merozoite contacts which result in invasion in untreated parasites and echinocytosis in the invasion-blocked cytD treated parasites. (C) The time from initial contact to the start of echinocytosis is shown for untreated vs. cytD treated parasites, irrespective of invasion.

Fig 5. Immediately prior to invasion a Fluo-4 signal appears at the junction of the merozoite and host cell suggesting an opening between the erythrocyte cytoplasm and the merozoite.

(A) Selected images from videos of purified W2mef schizonts added to Fluo-4 AM-labelled erythrocytes with (right) or without (left) cytD treatment. Outlines of the merozoite boundaries are transposed onto the Fluo-4 images to give an indication of where the Fluo-4 signal is relative to the merozoite. For those without cytD treatment, in the left panel, the direction of invasion is shown with an arrow, or a circle with a dot if the invasion is going into the plane of the image. (B) A timeline to scale showing the average length of time for each stage of the invasion process for untreated W2mef schizonts invading Fluo-4-treated erythrocytes. Above the timeline the average length of the Fluo-4 signal is indicated by the bar, with minimum and maximum shown by the whiskers, and placed on the timeline relative to the start of invasion. (C) Mean and SD for each stage of invasion and the timing of the punctate calcium signals observed for untreated W2mef schizonts invading Fluo-4-treated erythrocytes. (D) Model showing reorientation and illustrating how an open junction between merozoite rhoptries containing Ca²⁺ and erythrocyte cytoplasm containing Fluo-4 could mix at the merozoite apex, indicating a permeabilization or opening of the erythrocyte membrane. (E) Percent of total merozoites with punctate calcium signals which cause echinocytosis or do not, and within these the proportion which invade.

Fig 6. A new model for merozoite invasion.

A diagram of an invading merozoite showing the order and function of the major receptor-ligand interactions studied here. Below this scheme are shown selected molecular interactions critical for invasion. This model incorporates for the first time a sequential hierarchy of distinct receptor-ligand steps during pre-invasion. We show the relationship of these interactions to erythrocyte deformation, which precedes normal successful invasion. Our data implies that rhoptry release is dependent on the interaction between PfRh5 and basigin, resulting in an opening in the erythrocyte membrane, and that external calcium is also important at this stage of invasion.