Plasmodium falciparum Adhesins Play an Essential Role in Signalling and Activation of Invasion into Human Erythrocytes

Abstract

The most severe form of malaria in humans is caused by the protozoan parasite Plasmodium falciparum. The invasive form of malaria parasites is termed a merozoite and it employs an array of parasite proteins that bind to the host cell to mediate invasion. In Plasmodium falciparum, the erythrocyte binding-like (EBL) and reticulocyte binding-like (Rh) protein families are responsible for binding to specific erythrocyte receptors for invasion and mediating signalling events that initiate active entry of the malaria parasite. Here we have addressed the role of the cytoplasmic tails of these proteins in activating merozoite invasion after receptor engagement. We show that the cytoplasmic domains of these type 1 membrane proteins are phosphorylated in vitro. Depletion of PfCK2, a kinase implicated to phosphorylate these cytoplasmic tails, blocks P. falciparum invasion of red blood cells. We identify the crucial residues within the PfRh4 cytoplasmic domain that are required for successful parasite invasion. Live cell imaging of merozoites from these transgenic mutants show they attach but do not penetrate erythrocytes implying the PfRh4 cytoplasmic tail conveys signals important for the successful completion of the invasion process.

Fig 1. The cytoplasmic tails of P. falciparum adhesins are phosphorylated in vitro.

(A) The cytoplasmic tail sequences of the EBL and PfRh protein families are shown. Putative phosphorylation sites for kinases as predicted by NetPhosK are shown with the predicted kinase above the residue and represent the prediction score of the algorithm above 0.5. Residues that have a predicted CK2 site are labelled with the corresponding amino acid number. (B) In vitro kinase assays of EBL and PfRh cytoplasmic tails with merozoite lysates. Autoradiograph of in vitro kinase assays ³²P with EBL and PfRh cytoplasmic tail sequences fused to GST. Merozoite lysates were used as a source of kinase activity and the coomasie brilliant blue stained gel of the individual fusion proteins is shown in the right panel. All proteins show labelling with radioactive phosphate except GST alone and the PfRh1 cytoplasmic tail. Molecular weight markers in kDa are shown on the left of the gel. (C) In vitro kinase assays of EBL cytoplasmic tails with a small panel of P. falciparum recombinant kinases. The first lane has only the recombinant cytoplasmic tail whereas subsequent lanes have both recombinant tail and kinase as labelled. Some of the EBL and PfRh tails are phosphorylated by PfCK2 but not by any other kinases in the panel. Migration of each recombinant protein is highlighted with an asterisk.

Fig 2. Identification of putative phosphosites and parasite kinase involved in modification of PfRh4 cytoplasmic tail.

(A) In vitro kinase assays of wildtype and putative phosphosite mutations in PfRh4 cytoplasmic domains. The amino acid sequence of the PfRh4 cytoplasmic tail is shown with serines and tyrosines highlighted with the residue number. Each potential phosphosite on the PfRh4 tail was individually mutated to alanine. The all 4 lane has mutations in S1667A, S1674A, Y1680A and Y1684A and the all 5 lane has S1652A, S1667A, S1674A, Y1680A and Y1684A putative kinase sites mutated. The phosphorylation signal was quantitated and adjusted for protein loading. The loading-adjusted mutant phosphorylation signals were divided by the wildtype and plotted as a percentage of the wildtype signal (Y-axis). Autoradiograph of proteins after incubation in the in vitro phosphorylation assay and Coomassie gel from which protein loading was quantitated are shown in lower panels. Lane labels (X-axis) denote residues mutated to alanine. Mean percentage of wildtype phosphorylation +1 standard error of the mean are displayed. Data was averaged from four experiments performed on separate days. (B) Dosage-response curve for PfRh4 tail phosphorylation by merozoite lysate in the presence of increasing concentrations of the CK2 inhibitor TBB. PfRh4 tail phosphorylation was quantitated after incubation in in vitro phosphorylation assay with TBB. The phosphorylation signal for each condition was adjusted to reflect the average amount of protein loaded across each condition, determined by densitometry of the Coomassie brilliant blue stained gel. Y-axis represents loading-adjusted phosphorylation signal as a percentage of phosphorylation in the presence of DMSO (control). Autoradiograph of wildtype GST-fused PfRh4 proteins after incubation in the in vitro phosphorylation assay. X-axis indicates the TBB concentration with which the phosphorylation assay was incubated or DMSO. (C) In vitro kinase assays of PfRh and EBL cytoplasmic tails. The phosphorylation signal was quantitated and adjusted for protein loading. Autoradiograph of proteins after incubation in the in vitro phosphorylation assay and Coomassie brilliant blue stained gel from which protein loading was quantitated are shown. Data was averaged from four experiments performed on separate days (right panel) and standard error of the mean is shown. The following sites were mutated: EBA140 (S1159A, S1168A, T1173A), EBA175 (T1466A, mut A) and (S1489A, mut B) and in combination (mut A and B), EBA181 (S1528A, S1553A, S1557A, T1564A), PfRh2a (S3128A) and PfRh2b (S3233).

Fig 3. The effect of PfCK2 knockdown on PfRh4 phosphorylation.

(A) PfCK2-HA-DD retains kinase activity. PfCK2-HA-DD was immuno-precipitated from parasite lysate using an anti-HA antibody. In vitro kinase assay was performed using the immune-precipitatied eluate with the absence or presence of heparin. (B) Time course PfCK2 expression in 3D7-PfCK2DD parasites in the presence and absence of Shield-1. Shield-1 was removed at the ring stage of a synchronous parasite population, Day 0 (5% parasitemia). Absence of Shield-1 resulted in the progressive down-regulation of PfCK2. Parasite proteins were extracted using saponin lysis and analysed by western blotting for the presence of HA-tagged PfCK2 protein. HSP70 served as a loading control for normalisation. Densitometry analysis was performed and shown in bottom graph. (C) Growth assay of PfCK2-HA-DD in the presence and absence of Shield-1 (n = 2, error bars are standard error of the mean (SEM). Identical experiments were performed on a second cloned parasite line and similar results were obtained (n = 2). Parasitemia was quantified by counting Giemsa-stained smears. (D) Progression of PfCK2-HA-DD from late schizont to ring formation with and without Shield-1. Distinct ring, trophozoite and schizont populations were monitored using Sybr Green staining as the parasite progressed from 44 to 50 hours. One representative experiment is shown and was repeated 3 independent times. (E) Progression of PfCK2-HA-DD from late schizont to ring formation with and without Shield-1. Distinct schizont, late segmenter and ring populations were monitored using light microscopy and Giemsa-stained slides as the parasite progressed hourly from 42 to 48 hours. This experiment was performed independently twice and error bars shown are standard error of the mean (SEM).

Fig 4. Substitution of the PfRh4 cytoplasmic tail results in a defect in merozoite invasion.

(A) Schematic of the substitution of PfRh4 cytoplasmic tail with cytoplasmic domains from EBA-175 and AMA1. The amino acid sequence of each respective tail region is shown at the bottom. (B) Expression of the transgenic PfRh4 lines containing a substitution in the cytoplasmic tails. PfRh4 expression was seen in W2mefΔ175 but not in the W2mefΔRh4 or parental W2mef strain as expected. The difference in protein migration between PfRh4-175tail and PfRh4-AMA1 tail is likely due to the difference in acidic residues between the various cytoplasmic tails. The molecular weight marker is labelled on the left of the panel (kDa). (C) Growth assays of the PfRh4 substituted cytoplasmic tail transgenic lines. Parasitaemia was measured in neuraminidase-treated, and untreated erythrocytes after every 48 hours incubation (labelled as cycles). W2mef∆175 and Rh4-WT tail expresses PfRh4 and can invade neuraminidase-treated erythrocytes whereas W2mef∆PfRh4 does not express PfRh4 under these experimental conditions are not able to invade neuraminidase-treated erythrocytes. W2mef shows an ability to invade neuraminidase-treated erythrocytes upon selection on these cells. Both Rh4-175tail and Rh4-AMA1tail were not able to fully grow in neuraminidase-treated erythrocytes. The parasite lines used in this experiment are displayed on the X-axis. The y-axis represents parasitaemia of neuraminidase-treated erythrocytes as a percentage of parasitaemia of the same line grown on untreated erythrocytes. Error bars represent +1 standard error of the mean. Assay performed three times on separate days, each in triplicate.

Fig 5. Specific amino acid residues in the cytoplasmic domain are essential for PfRh4 function.

(A) Localization of PfRh4 and PfRh2 in Rh4-WT tail, RH4-AMA1tail, Rh4-mut4tail and RH4-mut5tail lines as detected using anti-PfRh4 monoclonal and anti-PfRh2 polyclonal antibodies. Parasite nuclei were stained with DAPI. (B) Expression of PfRh4 and PfRh2 in W2mefΔ175 and all transgenic lines as detected by anti-PfRh4 and anti-PfRh2 antibody. All transgenic lines migrated as a slightly larger doublet compared to PfRh4 in W2mefΔ175, consistent with the addition of the hexa-histidine tag at the C-terminus of the protein. (C) Growth assays of transgenic lines with four and five mutations at serine and tyrosine amino acid residues in the PfRh4 cytoplasmic tail. (D) Growth assays of single mutations in the PfRh4 cytoplasmic tail. (E) Growth assays of double mutations in the PfRh4 cytoplasmic tail. In all three panels, parasitaemia was measured in neuraminidase-treated, and untreated erythrocytes after every 48 hours incubation (labelled as cycles). The parasite lines used in this experiment are displayed on the X-axis. The y-axis represents parasitaemia of neuraminidase-treated erythrocytes as a percentage of parasitaemia of the same line grown on untreated erythrocytes. Error bars represent +1 standard error of the mean. Assay performed three times on separate days, each in triplicate.

Fig 6. Live Imaging of Rh4-mut tail variant show that parasites are unable to efficiently invade neuramindase—treated erythrocytes.

(A) Graphical interpretation of the video microscopy of live merozoites expressing modified PfRh4 tails attempting to invade NM treated and untreated erythrocytes. (B) Merozoite invasions per schizont rupture in NM-treated and untreated erythrocytes. Ruptures = the number of schizont ruptures that were imaged and followed for merozoite invasion. (C) The percentage of merozoite contacts with erythrocytes lasting ≥0.25s that leads to successful invasion. Shown is the average (mean and standard deviation) of the percentages of invasions per number of contacts observed for each schizont rupture. (D) Deformation scores of merozoites contacting erythrocytes in NM-treated and non-treated erythrocytes. Shown are the percentages of each deformation score from the total number of contacts. The total number of events (contact between a merozoite and RBC) counted for each experimental condition is shown. Statistics: B and C, Mann-Whitney test where *p<0.05, **p>0.01, **p<0.01; D, Chi square test.