Title: A pan‐apicomplexan phosphoinositide‐binding protein acts in malarial microneme exocytosis
Abstract: Scientific Report16 May 2019free access Transparent process A pan-apicomplexan phosphoinositide-binding protein acts in malarial microneme exocytosis Zeinab Ebrahimzadeh Centre de Recherche en Infectiologie, CRCHU de Québec-Université Laval, Québec, QC, Canada Search for more papers by this author Angana Mukherjee Centre de Recherche en Infectiologie, CRCHU de Québec-Université Laval, Québec, QC, Canada Search for more papers by this author Marie-Ève Crochetière Centre de Recherche en Infectiologie, CRCHU de Québec-Université Laval, Québec, QC, Canada Search for more papers by this author Audrey Sergerie Centre de Recherche en Infectiologie, CRCHU de Québec-Université Laval, Québec, QC, Canada Search for more papers by this author Souad Amiar Department of Medicinal Chemistry and Molecular Pharmacology and the Purdue Institute of Inflammation, Immunology and Infectious Disease, Purdue University, West Lafayette, IN, USA Search for more papers by this author L Alexa Thompson Division of Infectious Disease, Department of Medicine, Faculty of Medicine and Dentistry, University of Alberta, Edmonton, AB, Canada Search for more papers by this author Dominic Gagnon Centre de Recherche en Infectiologie, CRCHU de Québec-Université Laval, Québec, QC, Canada Search for more papers by this author David Gaumond Centre de Recherche en Infectiologie, CRCHU de Québec-Université Laval, Québec, QC, Canada Search for more papers by this author Robert V Stahelin Department of Medicinal Chemistry and Molecular Pharmacology and the Purdue Institute of Inflammation, Immunology and Infectious Disease, Purdue University, West Lafayette, IN, USA Search for more papers by this author Joel B Dacks Division of Infectious Disease, Department of Medicine, Faculty of Medicine and Dentistry, University of Alberta, Edmonton, AB, Canada Search for more papers by this author Dave Richard Corresponding Author [email protected] orcid.org/0000-0003-2269-6240 Centre de Recherche en Infectiologie, CRCHU de Québec-Université Laval, Québec, QC, Canada Search for more papers by this author Zeinab Ebrahimzadeh Centre de Recherche en Infectiologie, CRCHU de Québec-Université Laval, Québec, QC, Canada Search for more papers by this author Angana Mukherjee Centre de Recherche en Infectiologie, CRCHU de Québec-Université Laval, Québec, QC, Canada Search for more papers by this author Marie-Ève Crochetière Centre de Recherche en Infectiologie, CRCHU de Québec-Université Laval, Québec, QC, Canada Search for more papers by this author Audrey Sergerie Centre de Recherche en Infectiologie, CRCHU de Québec-Université Laval, Québec, QC, Canada Search for more papers by this author Souad Amiar Department of Medicinal Chemistry and Molecular Pharmacology and the Purdue Institute of Inflammation, Immunology and Infectious Disease, Purdue University, West Lafayette, IN, USA Search for more papers by this author L Alexa Thompson Division of Infectious Disease, Department of Medicine, Faculty of Medicine and Dentistry, University of Alberta, Edmonton, AB, Canada Search for more papers by this author Dominic Gagnon Centre de Recherche en Infectiologie, CRCHU de Québec-Université Laval, Québec, QC, Canada Search for more papers by this author David Gaumond Centre de Recherche en Infectiologie, CRCHU de Québec-Université Laval, Québec, QC, Canada Search for more papers by this author Robert V Stahelin Department of Medicinal Chemistry and Molecular Pharmacology and the Purdue Institute of Inflammation, Immunology and Infectious Disease, Purdue University, West Lafayette, IN, USA Search for more papers by this author Joel B Dacks Division of Infectious Disease, Department of Medicine, Faculty of Medicine and Dentistry, University of Alberta, Edmonton, AB, Canada Search for more papers by this author Dave Richard Corresponding Author [email protected] orcid.org/0000-0003-2269-6240 Centre de Recherche en Infectiologie, CRCHU de Québec-Université Laval, Québec, QC, Canada Search for more papers by this author Author Information Zeinab Ebrahimzadeh1, Angana Mukherjee1,‡, Marie-Ève Crochetière1,‡, Audrey Sergerie1,‡, Souad Amiar2, L Alexa Thompson3, Dominic Gagnon1, David Gaumond1, Robert V Stahelin2, Joel B Dacks3 and Dave Richard *,1 1Centre de Recherche en Infectiologie, CRCHU de Québec-Université Laval, Québec, QC, Canada 2Department of Medicinal Chemistry and Molecular Pharmacology and the Purdue Institute of Inflammation, Immunology and Infectious Disease, Purdue University, West Lafayette, IN, USA 3Division of Infectious Disease, Department of Medicine, Faculty of Medicine and Dentistry, University of Alberta, Edmonton, AB, Canada ‡These authors contributed equally to this work *Corresponding author. Tel: +1 418 525 4444 ext 47975; Fax: +1 418 654 2715; E-mail: [email protected] EMBO Rep (2019)20:e47102https://doi.org/10.15252/embr.201847102 PDFDownload PDF of article text and main figures. Peer ReviewDownload a summary of the editorial decision process including editorial decision letters, reviewer comments and author responses to feedback. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info Abstract Invasion of human red blood cells by the malaria parasite Plasmodium falciparum is an essential step in the development of the disease. Consequently, the molecular players involved in host cell invasion represent important targets for inhibitor design and vaccine development. The process of merozoite invasion is a succession of steps underlined by the sequential secretion of the organelles of the apical complex. However, little is known with regard to how their contents are exocytosed. Here, we identify a phosphoinositide-binding protein conserved in apicomplexan parasites and show that it is important for the attachment and subsequent invasion of the erythrocyte by the merozoite. Critically, removing the protein from its site of action by knock sideways preferentially prevents the secretion of certain types of micronemes. Our results therefore provide evidence for a role of phosphoinositide lipids in the malaria invasion process and provide further insight into the secretion of microneme organelle populations, which is potentially applicable to diverse apicomplexan parasites. Synopsis PfPH2 is a pan-apicomplexan phosphoinositide-binding protein important to attach and invade a red blood cell by the malaria parasite Plasmodium falciparum. It is implicated in the exocytosis of the micronemes secretory organelles. PfPH2 is a phosphoinositide binding protein. PfPH2 is important for the attachment to and invasion of the red blood cell. PfPH2 is important for the secretion of PfEBA-containing micronemes. PfPH2 is conserved throughout apicomplexan parasites. Introduction With more than 216 million cases and 445,000 deaths in 2016, malaria still represents a devastating disease 1. The widespread occurrence of antimalarial drug resistance and the lack of a commercialized vaccine highlight the need for novel therapeutics. Plasmodium spp., the etiological agents of the disease, are obligate intracellular parasites, and their invasion of human red blood cells (RBCs) is an essential part of their life cycle. The invasion process debuts with the initial recognition of the erythrocyte membrane by merozoite surface proteins 2-4 after which the merozoite reorientates so that its apical tip becomes juxtaposed to the RBC membrane. Tight attachment of the parasite then occurs through the binding of parasite invasion ligands to RBC surface receptors. These ligands are the erythrocyte binding-like (EBAs) and reticulocyte binding-like (RHs) proteins to their cognate receptors 5-8. Subsequently, a tight junction is formed through an interaction between the apical membrane antigen 1 (AMA1) and the rhoptry neck (RON) complex 9-11. Through an acto-myosin molecular motor, the parasite pulls itself into a parasitophorous vacuole in which it will reside 12-14. Several of the effector proteins involved in the process of merozoite invasion are stored in the apical complex organelles and released in a controlled fashion 15. How apical organelles are secreted by the merozoite is poorly known but evidence suggests that calcium and cGMP signaling are implicated 16, 17. Indeed, studies have shown that microneme discharge through the activation of calcium-dependent protein kinases (CDPKs) 18, 19 and the cyclic GMP-dependent protein kinase (PKG) 20, 21 is required for egress of merozoites from the schizont. Of interest, it was proposed that the disparate roles of PKG throughout the malaria parasite life cycle could potentially be explained by its regulation of phosphoinositide metabolism and its effect on calcium signaling and potentially vesicular trafficking 17. Exposure to low potassium levels as found in human plasma leads to a rise in intracellular calcium that then triggers the secretion of the micronemal proteins PfAMA1 and PfEBA175 and the subsequent interaction of the latter with glycophorin A on the RBC surface then results in the exocytosis of the rhoptries 22, 23. The interaction of PfRH1 with its as yet unknown receptor also results in an increase in intracellular calcium and secretion of PfEBA175 24. The ability of the PLC inhibitor U73122 to abrogate microneme secretion suggests that the P. falciparum PLC homologue is implicated in the process 22. Recent work in the related apicomplexan parasite Toxoplasma gondii further suggested that recognition of phosphatidic acid produced through the action of TgPI-PLC by an acylated Pleckstrin homology protein (TgAPH) present on the surface of the micronemes led to their exocytosis and parasite egress. The authors further showed that recombinantly expressed P. falciparum APH also bound to PA but whether it plays an equivalent role in microneme exocytosis is unknown 25, 26. Finally, the snare-like C2 domain-containing protein PfDOC2.1 was shown to be required for the secretion of the micronemal protein PfEBA175 and the rhoptry neck protein PfRH2a 27, 28. Until recently and despite early evidence suggesting that the P. falciparum micronemes were composed of heterogenous populations with specific functions in egress and/or invasion 29, most studies extrapolated results obtained while studying one micronemal protein (most often PfAMA1) to apply to all the others. Of interest, this heterogeneity of micronemes is also conserved in T. gondii 30, 31. A few studies have now started to take this into consideration. For example, recent results have shown that PfCDPK1 knockdown parasites have no defect in egress but cannot invade erythrocytes due to a specific defect in the secretion of PfEBA175-containing micronemes but not of PfAMA1 micronemes 32. Absalon and colleagues revealed that the signaling cascade containing PfPKG and PfCDPK5 led to the secretion of PfAMA1 on the surface of merozoite before egress but not of PfEBA175 which remains in the micronemes and led them to suggest the inclusion of additional subsets of egress-specific micronemes 18. The mechanisms underlying this differential exocytosis are currently unknown. Intriguingly, despite representing only a minor fraction of the total lipids of eukaryotic membranes, phosphoinositides (PIPs) are critical components involved in a variety of cellular processes and recent work has shown that it was also the case for apicomplexan parasites (reviewed in 33). More specifically for P. falciparum, roles for PIPs have been shown in cytokinesis and merozoite formation 34, apicoplast biogenesis and inheritance 35-37, hemoglobin endocytosis 38, merozoite egress 17, gametocyte activation 39-41, and ookinete motility 17, the latter two occurring in the mosquito vector, and finally in resistance to artemisinin 42, 43. In line with these varied functions, the determination of the subcellular distribution of several species of PIPs in the malaria parasite asexual erythrocytic stages showed localizations to structures such as the Golgi apparatus, the plasma membrane, the food vacuole membrane, the endoplasmic reticulum, and the apicoplast 34, 35, 44, 45. While hydrolysis of PI(4,5)P2 by the Toxoplasma gondii PI-phospholipase C is critical for the invasion of this parasite (see below, 25), a direct role for PIPs in the malaria merozoite invasion process has not yet been described. We here identify a P. falciparum PH domain-containing protein that is important for the secretion of micronemes containing PfEBAs. This protein has a relaxed phosphoinositide-binding ability and we show, using knock sideways (KS), that it is required for merozoite attachment and invasion. Furthermore, we show bioinformatically that this protein is present in diverse apicomplexans suggesting that it plays a role in pathogenesis beyond Plasmodium. Our results therefore provide further insight into the secretion of micronemal populations. Results and Discussion The Pleckstrin homology domain of PF3D7_1337700 has a relaxed phosphoinositide-binding specificity As part of our efforts to investigate potential roles for PIPs in the invasion process, we identified PF3D7_1337700 (www.plasmodb.org), a putative PIP-binding protein containing a PH domain (Fig 1A). Less than 10% of the characterized PH domains possess the ability to bind PIPs but a common feature of PIP-binding PH domains is the presence of a basic sequence motif KXn (K/R)XR involved in the binding to the head group of the inositol moiety 46. Inspection of the PF3D7_1337700 PH domain sequence revealed that such a motif was present (84KANIFYIYKLR94, Fig 1A). To determine whether the PH domain had the capacity to bind to PIPs, we recombinantly produced the WT PH domain and a version where residues K84, K92, and R94 were mutated to alanines, fused to an N-terminal glutathione-S-transferase tag. Coomassie staining of the purified proteins revealed a band at the expected size of 38 kDa along with a 26-kDa degradation product likely corresponding to GST alone (Appendix Fig S1). Incubation of the GST-WT PH domain with PIP Strips showed that the protein interacts significantly with PI(3)P, whereas the well-established PLCδ PH domain bound PI(4,5)P2 47 (Fig 1B). The WT PH domain also showed some residual interaction with PI(4)P, PI(5)P (Fig 1B). The PH domain triple mutant was used to confirm the specificity of this domain to phosphoinositides (PIPs). Lipid overlay showed the ability of the PH domain mutant to interact the same PIPs than the wild type with higher affinity to PI(3)P (Fig 1B). Figure 1. PF3D7_1337700 is a phosphoinositide-binding protein with a relaxed specificity Schematic of PF3D7_1337700 showing the PH domain with the conserved PIP-binding motif. Lipid blots showing that the WT PH domain and triple mutant PH domain bind several species of PIPs. A negative control (GST) and a positive control (PLCδ-PH) are also shown for respective lipid binding. Liposome-binding assays of GST-tagged WT and triple mutant PH domains. GST-tagged PfPH2-PH proteins (500 ng) were incubated with 50 μM liposomes composed of POPC:POPE, POPC:POPE:POPS, or liposomes containing 5% molar ratio of one of seven PIP species or DPPA (POPC:POPE:POPS:DPPA or POPC:POPE:POPS:PIPs). kDa indicates molecular weight, P: pellet fraction, SN: supernatant fraction. SPR analysis of the WT PfPH2-PH domain and PH domain triple mutant demonstrates binding to lipid vesicles (POPC:POPE) containing 5 mol% PI(3)P. The response values shown (determined by subtracting the binding signal from control lipid vesicles) were plotted versus PH domain concentration to determine the apparent affinity (Kd) of vesicle binding. Download figure Download PowerPoint PIP strips contain lipids spotted on nitrocellulose, which cannot recapitulate a membrane bilayer. To determine whether the PH domain and triple mutant could interact with PIPs in a membrane bilayer, we employed a liposome-binding assay using a 5% molar ratio of PIPs or DPPA (phosphatidic acid) containing liposomes. The liposome-binding assay indicated the selective binding of WT PH domain to the phosphoinositol monophosphate species. Indeed, as shown in Fig 1C, about 20% of the protein was found in the pellet fraction of PI(3)P, PI(4)P, or PI(5)P. The WT PH domain also exhibited significant binding to PI(4,5)P2 and a small amount of detectable binding to PI(3,4,5)P3 and PA containing vesicles. However, no binding of WT PH domain was observed for two other phosphoinositol bisphosphate containing membranes (PI(3,4)P2 and PI(3,5)P2) (Fig 1C, top panel). In contrast, the triple mutation of the PH domain showed a clear loss of the ability to bind liposomes containing different PIPs (Fig 1C, bottom panel). Taken together, these data indicate that the WT PH domain can interact robustly with several PIP species in lipid bilayers, whereas the triple mutant had a greatly diminished ability to interact with PIP-containing membranes. Since the liposome-binding assay showed that the triple mutant lost the ability of a stable binding to specific PIP species, this suggests that the basic sequence motif identified (K84, K92, and R94) may be involved in the binding to the PIP head group in the lipid bilayer. To confirm this hypothesis, we performed surface plasmon resonance (SPR) analysis to determine the binding affinity of both WT and the PH domain triple mutation to PI(3)P-containing lipid vesicles. We employed lipid vesicles on the surface of a L1 sensor chip. The SPR assay confirmed the ability of the WT PH domain to bind lipid vesicles containing PI(3)P with an apparent disassociation constant (Kd) of 630 nM (Fig 1D and Appendix Fig S2). In contrast, the PH domain triple mutant had weak binding to PI(3)P-containing vesicles and we were not able to determine an apparent Kd for the triple mutant (Fig 1D and Appendix Fig S2). The apparent PI(3)P membrane binding affinity of the WT PH domain is on par with other well-characterized PIP-binding domains 48. Promiscuous PIP binding has been described for 67% of yeast PH domains, and their specific subcellular distribution requires coincidence detection of additional factors such as another protein or membrane curvature for example 49. These results demonstrate that PF3D7_1337700 is a true PIP-binding protein and we renamed it PfPH2 following the uncharacterized PfPH1 named after a T. gondii PI(3,5)P2-binding protein 50. PfPH2 is likely essential for the asexual erythrocytic cycle To further characterize PfPH2, we endogenously tagged its C terminus with GFP by single cross-over recombination using the recently developed selection-linked integration (SLI) strategy 51. To allow the functional analysis of PfPH2 by KS (see below), a double FK506 binding protein domain (2xFKBP) tag was also appended 51 (Fig EV1A). Proper integration of the vector and the absence of a WT allele were verified by polymerase chain reaction (PCR) demonstrating that we had successfully tagged the pfph2 gene (Fig EV1B). Time course analysis of PfPH2-2xFKBP-GFP expression by Western blot using an anti-GFP antibody on parasite protein extracts taken throughout the asexual erythrocytic cycle (from the ring through to the schizont stage) revealed a single band at the expected size of around 133 kDa for the PfPH2-2xFKBP-GFP fusion protein in schizont stage parasites. An antibody against the constitutive protein PfHSP70 was used as staging control (Fig EV1C). Immunofluorescence assays (IFA) of PfPH2-2xFKBP-GFP parasites showed a faint punctate signal in late schizont stages that did not colocalize with any of the markers investigated (micronemes: PfAMA1, PfEBA175, and PfEBA140 (Fig EV2A); rhoptry bulb: PfRAP1 or neck: PfRON4, RH1, RH4, and RH5 (Fi EV2B); dense granules: PfRESA (Fig EV2C); or the Golgi apparatus: PfERD2 (Fig EV2D). To try to get a less crowded view than with schizonts, IFAs were also performed on free merozoites recently egressed. Again, no strong colocalization could be seen between PfPH2 and any of the markers (Fig EV3A and B). However, the PfPH2 signal often seemed to be more apical (Fig EV3Aii and iii and B). To look at this in a more quantitative manner, we calculated the distance between the farthest edge of the DAPI and PfPH2, PfRON4, and PfEBA175 and used this to infer how apical a protein potentially is, i.e., the bigger the distance between the marker and the DAPI, the more apical a protein likely is. These data show that PfPH2 is farther from the DAPI than PfEBA175 (91.34 ± 1.78 versus 75.60 ± 1.72 pixels, Fig 2Ai and B) and potentially more than PfRON4 (90.83 ± 2.83 versus 83.19 ± 2.49 pixels) though the latter difference was not statistically significant (Fig 2Aii and B). PfRON4 is a well-described marker of the rhoptry neck so our results suggest that PfPH2 localizes to a structure close to the apical tip of the merozoite. Click here to expand this figure. Figure EV1. Generation of the PfPH2-2xFKBP-GFP parasite line Schematic showing the tagging strategy by single cross-over recombination using SLI. PCR on parasite genomic DNA showing the proper integration of the tagging vector at (5′ junction: primers P1 and P3, 3′ junction: primers P2 and P4) and the disappearance of the WT allele (primers P1 and P4) in all 3 SLI attempts. Time course of expression of parasite protein extracts taken throughout the erythrocytic cycle shows that PfPH2 is expressed in schizonts. PfHSP70 is used as a control for a constitutive protein. Cartoon shows the time points in hours post-invasion (hpi). 1: 8 hpi, 2: 30 hpi, 3: 36 hpi, 4: 42 hpi. Download figure Download PowerPoint Click here to expand this figure. Figure EV2. Localization of PfPH2 in schizonts A–D. IFA showing that PfPH2 does not colocalize with markers of the micronemes: PfAMA1, PfEBA175, and PfEBA140) (A), of the rhoptries: PfRAP1, PfRON4, PfRH1, PfRH4, and PfRH5 (B), of the dense granules: PfRESA (C) or of the Golgi apparatus: PfERD2 (D). Scale bar represents 5 μm. Blue: DAPI-stained nucleus. BF: bright field. Download figure Download PowerPoint Click here to expand this figure. Figure EV3. Localization of PfPH2 in free merozoites A, B. IFA showing that PfPH2 does not colocalize with markers of the micronemes: PfAMA1 (secreted on the parasite surface in free merozoites), PfEBA175, and PfEBA140 (A) or of the rhoptry markers PfRON4, PfRAP1, PfRH1, and PfRH5 (B). Scale bar represents 5 μm. Blue: DAPI-stained nucleus. Download figure Download PowerPoint Figure 2. PfPH2 localizes close to the apical tip IFA showing the distance between the farthest edge of the DAPI and PfPH2 and (Ai) PfEBA175 and (Aii) PfRON4. Two images are shown for EBA175 to demonstrate how much variety can be observed between different cells. Green: PfPH2. Red: EBA175 (Ai) or RON4 (Aii). Blue: DAPI-stained nucleus. Scale bar: 0.5 μm. Quantification of the distances reveals that the distance between PfPH2 and the DAPI is significantly bigger than PfEBA175 and the latter but not with PfRON4. ****P < 0.0001. ns: non-significant. One-way ANOVA. Horizontal lines: median. Box limits: 25th to 75th percentile. Whiskers: min to max values. Error bars: standard error of the mean. Download figure Download PowerPoint To investigate the essentiality of PfPH2 for the asexual erythrocytic cycle, we first tried to knock out (KO) its gene by single cross-over recombination using the SLI for targeted gene disruption strategy (SLI-TGD) 51. The fact that we could never detect proper integration of the vector by PCR and the persistence of the WT allele in three KO attempts suggests that PfPH2 might be required for the asexual stage growth (unpublished observations). This is further supported by a recent whole-genome functional screen using saturation mutagenesis by the piggyBac transposon 52. To gain insight into the role of PfPH2 in the asexual blood stages, we performed KS, a strategy that allows the conditional removal of a protein of interest from its site of action 53. To do so, we transfected the PfPH2-2xFKPB-GFP parasite line with an episome expressing a nuclear mislocalizer consisting of a triple nuclear localization signal fused to a double FKBP12-rapamycin binding domain and the cherry fluorescent protein (3xNLS-2xFRB-cherry) 51. We next tested the ability of the mislocalizer to translocate PfPH2-2xFKPB-GFP to the nucleus in the presence of rapamycin (Rapa), thus removing it from its normal site of action. As seen in Fig 3Ai, in absence of Rapa, the mislocalizer colocalizes with the DAPI-stained nucleus while PfPH2-2xFKPB-GFP shows its normal punctate pattern. When adding Rapa at the ring stage, before the expression of PfPH2 is turned on, and letting the parasites mature to late schizonts, a substantial proportion of the GFP signal was now observed in the nucleus instead of the apical foci (Fig 3Aii). We quantified the levels of KS by calculating Pearson's correlation coefficient for PfPH2-2xFKPB-GFP versus the mislocalizer and also versus the DAPI staining, in the presence and absence of Rapa. Our results show that the addition of Rapa does lead to a significant increase in the colocalization between GFP and both the mislocalizer (0.24 ± 0.02 versus 0.58 ± 0.02) and the DAPI (0.21 ± 0.01 versus 0.50 ± 0.01) at the population level though there is some variability when looking at individual cells (Appendix Fig S3). This shows that we succeeded in performing KS on PfPH2. Figure 3. PfPH2 is important for attachment to and subsequent invasion of the red blood cell Live-cell microscopy showing that in the presence of 250 nM Rapa, PfPH2-2xFKBP-GFP (GFP) is translocated from the apical pole to the nucleus in the presence of the nuclear mislocalizer. Scale bar represents 5 μm. Blue: DAPI-stained nucleus. BF: bright field. (Bi) Growth curve analysis showing that the KS of PfPH2 severely decreases the asexual replication of the parasite. The PfPH2-GFP parasite line (black square) not transfected with the nuclear mislocalizer is used as a control to show that the reduced growth is dependent on the presence of both Rapa and the mislocalizer. Full line: −Rapa. Dashed line: +Rapa. Mean ± SEM of three biological replicates is shown. (Bii) Data from (Bi) represented as the percentage of growth of parasites incubated with Rapa compared to their control without Rapa. Mean ± SEM of three biological replicates is shown. Time course of schizont rupture (red) and subsequent new ring formation (blue) showing that the PfPH2 KS line has no egress defect but an important decrease in the formation of new rings. Results from one experiment representative of three biological replicates are shown. Full line: −Rapa. Dashed line: +Rapa. Quantification of the number of merozoites formed per schizont. n = 4 biological replicates with 20 schizonts counted per condition. n.s: non-significant. Unpaired t-test. P = 0.1524. Error bars: Standard error of the mean. Merozoite attachment and invasion are decreased in the PfPH2 KS line. Attachment was measured by incubating parasites with 1 μM cytochalasin D. CytD: cytochalasin D. n = 3 biological replicates for both attachment and invasion assays. One-way ANOVA followed by Fisher's LSD test. Error bars: Standard error of the mean. Download figure Download PowerPoint We next determined the effect of the PfPH2 KS on the ability of the parasite to proliferate. To do this, tightly synchronous PfPH2-2xFKPB-GFP+ 3xNLS-2xFRB-cherry (PfPH2-GFP+ mislocalizer) parasites were incubated with or without Rapa and growth was monitored over two cycles. We first checked the effect of adding Rapa at different times and found that growth inhibition was maximal when the compound was added at the ring stage and that adding it shortly before egress decreased the effect by around 50% (unpublished observations). We hypothesize that this is due to the fact that PfPH2 is potentially strongly attached to an apical structure so that it cannot easily be "extracted" by the mislocalizer. For efficient mislocalization to occur, PfPH2 would need to be captured as it is synthesized. Based on this, all subsequent experiments were performed with Rapa added at the ring stage. As a control, we used the PfPH2-2xFKPB-GFP without the mislocalizer (PfPH2-GFP). As shown in Fig 3Bi and ii, the KS resulted in an around 65% decrease in parasitemia over one reinvasion cycle and up to 86% after the second reinvasion cycle while Rapa had no effect on the control line without mislocalizer. This suggests that PfPH2 is required for optimal proliferation of asexual stages. Having confirmed that the KS was dependent on the presence of the mislocalizer and that the concentration of Rapa used in our assays was not toxic in itself, the remaining experiments were only performed with the PfPH2+ mislocalizer line incubated with or without Rapa. We first looked at the integrity of various subcellular structures by IFA and could see no obvious difference in the PfPH2-GFP+ mislocalizer incubated with Rapa (referred to as the KS line onwards) (Appendix Fig S4). Next, to check whether the reduced parasitemia was due to a failure of the parasites to egress from the RBC, the number of schizonts and rings was followed over a 16-h period, at every 4 h, by Giemsa staining of parasite smears. As seen in Fig 3C and Appendix Fig S5, there was no difference in the evolution of schizont rupture between the KS and the control line; however, there was an important decrease in the number of rings formed. These data show that egress proceeds normally in the KS line. We next looked at whether the KS line produced fewer merozoites per schizont by counting DAPI-stained cells by fluorescence microscopy but again, no difference was seen (Fig 3D) which suggests that a defect in merozoite invasion was potentially the cause of the reduced parasitemia. To directly address this, invasion assays were performed with