Title: Transmission of the malaria parasite requires ferlin for gamete egress from the red blood cell
Abstract: Cellular MicrobiologyVolume 21, Issue 5 e12999 RESEARCH ARTICLEFree Access Transmission of the malaria parasite requires ferlin for gamete egress from the red blood cell Klara Obrova, Klara Obrova Center for Infectious Diseases, Parasitology Unit, Heidelberg University Hospital, Heidelberg, GermanySearch for more papers by this authorMarek Cyrklaff, Marek Cyrklaff Center for Infectious Diseases, Parasitology Unit, Heidelberg University Hospital, Heidelberg, GermanySearch for more papers by this authorRoland Frank, Corresponding Author Roland Frank [email protected] Center for Infectious Diseases, Parasitology Unit, Heidelberg University Hospital, Heidelberg, Germany Correspondence Roland Frank, Center for Infectious Diseases, Heidelberg University Hospital, Parasitology Unit, Im Neuenheimer Feld 324, Heidelberg 69120, Germany. Email: [email protected] Gunnar R. Mair, Department of Biomedical Sciences, Iowa State University, 2008 Vet Med, Ames, IA 50011. Email: [email protected] Ann-Kristin Mueller, German Center for Infectious Diseases (DZIF), Universitätsklinikum Heidelberg, Heidelberg 69120, Germany. Email: [email protected] for more papers by this authorGunnar R. Mair, Corresponding Author Gunnar R. Mair [email protected] Department of Biomedical Sciences, Iowa State University, Ames, Iowa, USA Correspondence Roland Frank, Center for Infectious Diseases, Heidelberg University Hospital, Parasitology Unit, Im Neuenheimer Feld 324, Heidelberg 69120, Germany. Email: [email protected] Gunnar R. Mair, Department of Biomedical Sciences, Iowa State University, 2008 Vet Med, Ames, IA 50011. Email: [email protected] Ann-Kristin Mueller, German Center for Infectious Diseases (DZIF), Universitätsklinikum Heidelberg, Heidelberg 69120, Germany. Email: [email protected] for more papers by this authorAnn-Kristin Mueller, Corresponding Author Ann-Kristin Mueller [email protected] orcid.org/0000-0001-7364-0090 Center for Infectious Diseases, Parasitology Unit, Heidelberg University Hospital, Heidelberg, Germany German Center for Infectious Diseases (DZIF), Universitätsklinikum Heidelberg, Heidelberg, Germany Correspondence Roland Frank, Center for Infectious Diseases, Heidelberg University Hospital, Parasitology Unit, Im Neuenheimer Feld 324, Heidelberg 69120, Germany. Email: [email protected] Gunnar R. Mair, Department of Biomedical Sciences, Iowa State University, 2008 Vet Med, Ames, IA 50011. Email: [email protected] Ann-Kristin Mueller, German Center for Infectious Diseases (DZIF), Universitätsklinikum Heidelberg, Heidelberg 69120, Germany. Email: [email protected] for more papers by this author Klara Obrova, Klara Obrova Center for Infectious Diseases, Parasitology Unit, Heidelberg University Hospital, Heidelberg, GermanySearch for more papers by this authorMarek Cyrklaff, Marek Cyrklaff Center for Infectious Diseases, Parasitology Unit, Heidelberg University Hospital, Heidelberg, GermanySearch for more papers by this authorRoland Frank, Corresponding Author Roland Frank [email protected] Center for Infectious Diseases, Parasitology Unit, Heidelberg University Hospital, Heidelberg, Germany Correspondence Roland Frank, Center for Infectious Diseases, Heidelberg University Hospital, Parasitology Unit, Im Neuenheimer Feld 324, Heidelberg 69120, Germany. Email: [email protected] Gunnar R. Mair, Department of Biomedical Sciences, Iowa State University, 2008 Vet Med, Ames, IA 50011. Email: [email protected] Ann-Kristin Mueller, German Center for Infectious Diseases (DZIF), Universitätsklinikum Heidelberg, Heidelberg 69120, Germany. Email: [email protected] for more papers by this authorGunnar R. Mair, Corresponding Author Gunnar R. Mair [email protected] Department of Biomedical Sciences, Iowa State University, Ames, Iowa, USA Correspondence Roland Frank, Center for Infectious Diseases, Heidelberg University Hospital, Parasitology Unit, Im Neuenheimer Feld 324, Heidelberg 69120, Germany. Email: [email protected] Gunnar R. Mair, Department of Biomedical Sciences, Iowa State University, 2008 Vet Med, Ames, IA 50011. Email: [email protected] Ann-Kristin Mueller, German Center for Infectious Diseases (DZIF), Universitätsklinikum Heidelberg, Heidelberg 69120, Germany. Email: [email protected] for more papers by this authorAnn-Kristin Mueller, Corresponding Author Ann-Kristin Mueller [email protected] orcid.org/0000-0001-7364-0090 Center for Infectious Diseases, Parasitology Unit, Heidelberg University Hospital, Heidelberg, Germany German Center for Infectious Diseases (DZIF), Universitätsklinikum Heidelberg, Heidelberg, Germany Correspondence Roland Frank, Center for Infectious Diseases, Heidelberg University Hospital, Parasitology Unit, Im Neuenheimer Feld 324, Heidelberg 69120, Germany. Email: [email protected] Gunnar R. Mair, Department of Biomedical Sciences, Iowa State University, 2008 Vet Med, Ames, IA 50011. Email: [email protected] Ann-Kristin Mueller, German Center for Infectious Diseases (DZIF), Universitätsklinikum Heidelberg, Heidelberg 69120, Germany. Email: [email protected] for more papers by this author First published: 29 December 2018 https://doi.org/10.1111/cmi.12999Citations: 11AboutSectionsPDF ToolsRequest permissionExport citationAdd to favoritesTrack citation ShareShare Give accessShare full text accessShare full-text accessPlease review our Terms and Conditions of Use and check box below to share full-text version of article.I have read and accept the Wiley Online Library Terms and Conditions of UseShareable LinkUse the link below to share a full-text version of this article with your friends and colleagues. Learn more.Copy URL Share a linkShare onFacebookTwitterLinkedInRedditWechat Abstract Ferlins mediate calcium-dependent vesicular fusion. Although conserved throughout eukaryotic evolution, their function in unicellular organisms including apicomplexan parasites is largely unknown. Here, we define a crucial role for a ferlin-like protein (FLP) in host-to-vector transmission of the rodent malaria parasite Plasmodium berghei. Infection of the mosquito vectors requires the formation of free gametes and their fertilisation in the mosquito midgut. Mature gametes will only emerge upon secretion of factors that stimulate the disruption of the red blood cell membrane and the parasitophorous vacuole membrane. Genetic depletion of FLP in sexual stages leads to a complete life cycle arrest in the mosquito. Although mature gametes form normally, mutants lacking FLP remain trapped in the red blood cell. The egress defect is rescued by detergent-mediated membrane lysis. In agreement with ferlin vesicular localisation, HA-tagged FLP labels intracellular speckles, which relocalise to the cell periphery during gamete maturation. Our data define FLP as a novel critical factor for Plasmodium fertilisation and transmission and suggest an evolutionarily conserved example of ferlin-mediated exocytosis. 1 INTRODUCTION Malaria is a global health problem with estimated 212 million cases and more than 400,000 deaths in 2016 (World Malaria Report 2017, World Health Organization). Plasmodium, the pathogenic single-celled protist causing malaria, alternates between the human host and the mosquito vector in a complex life cycle with diverse intracellular and extracellular stages adapted to different host cell types. The disease-causing asexual blood stage is followed by the sexual transmission stage represented by gametocytes. Male and female gametocytes form inside red blood cells (RBC) in the mammalian host, from sexually committed parasites, and remain developmentally quiescent until they are taken up during the insect's blood meal (Guttery, Roques, Holder, & Tewari, 2015). The environmental changes inside the mosquito midgut trigger a rapid developmental programme referred to as gametogenesis (Billker, Shaw, Margos, & Sinden, 1997). In response to the drop in temperature and the presence of xanthurenic acid, calcium ions are released from the parasite's intracellular stores, whereupon male gametocytes undergo three rounds of mitosis and produce eight motile, flagellated, and sperm-like microgametes (Brochet & Billker, 2016). Each female gametocyte develops into a single macrogamete inside the erythrocyte. The release of mature gametes from the RBC finally requires rupture of the two surrounding membranes (Kuehn & Pradel, 2010), namely, red blood cell membrane (RBCM) and parasitophorous vacuole membrane (PVM), delineating the intraerythrocytic niche of the parasite (Lingelbach & Joiner, 1998). Only the extracellular gametes are able to take part in fertilisation and form the infectious ookinete stage inside the mosquito midgut, thus securing parasite transmission. Gamete egress and specifically the permeabilisation of the membranes surrounding the parasite require the regulated trafficking and exocytosis of secretory vesicles that contain a diverse set of proteins (Bargieri et al., 2016; Deligianni et al., 2013; Kehrer, Singer, et al., 2016; Kehrer, Frischknecht, & Mair, 2016; Khan et al., 2005; Koning-Ward et al., 2008; Olivieri et al., 2015; Ponzi et al., 2009; Sinden, 1982; Talman et al., 2011). Despite the clear calcium-dependent nature of gamete maturation—the process involves PKG, CDPK1 and 4 and calcineurin (Billker et al., 2004; McRobert et al., 2008; Philip & Waters, 2015; Sebastian et al., 2012)—no egress molecule containing calcium binding domains that could directly link secretion of egress vesicles to the calcium levels has been described. In multicellular eukaryotes, proteins harbouring multiple C2 calcium sensor domains act as key regulators of exocytosis. An example of such a protein is Synaptotagmin I, which mediates presynaptic neurotransmitter release (Elferink, Peterson, & Scheller, 1993). In the hair cells of the inner ear, where synaptotagmin expression is absent, neurotransmitter release is mediated by otoferlin, a human ferlin protein with six (possibly seven) C2 domains (Pangršič, Reisinger, & Moser, 2012; Roux et al., 2006). Other members from the ferlin family also mediate exocytosis in a calcium-dependent manner and mutations in C2 domains typically lead to intracellular accumulation of secretory vesicles (Bansal & Campbell, 2004; Covian-Nares, Koushik, Puhl, & Vogel, 2010; Roux et al., 2006; Washington & Ward, 2006). These mutations are also linked to deafness (mutation of otoferlin; Yasunaga et al., 1999) or muscular dystrophy (mutation of dysferlin; Liu et al., 1998) in humans. The ferlins in the invertebrates Caenorhabditis elegans (Fer-1) and Drosophila melanogaster (Misfire) have roles in spermatogenesis (Achanzar & Ward, 1997; Ohsako, Hirai, & Yamamoto, 2003; Smith & Wakimoto, 2007; Washington & Ward, 2006). Recently, a Toxoplasma gondii ferlin has been linked to rhoptry secretion (Coleman et al., 2018). Although there are no synaptotagmin homologues encoded in the Plasmodium genome, ferlins comprise an ancient protein family conserved in unicellular organisms, including apicomplexan parasitic protists (Lek, Lek, North, & Cooper, 2010), but their function in these organisms remains largely unknown. Here, we provide genetic evidence that links the Plasmodium berghei ferlin-like protein (FLP) to exocytosis of egress vesicles and the disruption of the RBC and PV membranes required during the formation of free gametes and host-to-vector transmission. 2 RESULTS 2.1 flp, a member of the ferlin family, is expressed in asexual and sexual life cycle stages The P. berghei ferlin-like protein (FLP, PBANKA_1224400) contains six predicted C2 domains, a C-terminal transmembrane domain and a Fer domain (a ferlin-specific motif of unknown function; Staub, Fiziev, Rosenthal, & Hinzmann, 2004). These typical features cluster FLP together with ferlin (PBANKA_1319300) into the ferlin protein family (Figure 1a; Lek et al., 2010). The ferlin family is topologically highly conserved among various Plasmodium species, including Plasmodium falciparum (Figure 1a). The related apicomplexan Toxoplasma gondii harbours three ferlin proteins, of which TGME49_309420 is most similar to FLP (Figure 1a). Sequence similarities indicate that FLP function might be conserved among Plasmodium species and related apicomplexans (Figure S1). Ferlins are typically embedded in vesicular membranes via a single C-terminal transmembrane domain with the N-terminus of the protein facing the cytosol (Figure 1b; Lek, Evesson, Sutton, North, & Cooper, 2012). Figure 1Open in figure viewerPowerPoint FLP belongs to the ferlin protein family and is expressed in asexual and sexual blood stages. (a) Protein domain analysis of selected ferlins reveals highly conserved topology with multiple (4–6) C2 domains and a C-terminal transmembrane domain. Amino acid (AA) sequence identity in comparison to PBANKA_1224400 is shown in percent as calculated by Clustal. (b) Schematic of typical ferlin insertion into vesicles, where C2 domains face the cytosol. Domains are colour coded as in (a). (c) FLP::HA fusion protein (186 kDa) is detected by Western blot in schizont- and gametocyte-enriched samples. Lysates were loaded on gradient Bis-Tris Gel (Nu-PAGE); Hsp70 was used as loading control. (d) FLP::HA fusion protein is detected in schizonts and gametocytes and shows a vesicular staining. Parasites were PFA-fixed and stained against HA (green), DNA was visualised using Hoechst (blue). Confocal images, scale bars: 5 μm. (e) FLP::HA localises to membranous structures. Enriched and purified gametocytes were analysed using electron microscopy. Immunogold (anti-HA) signal localises to cytoplasmic vesicles. Magnified areas depict two regions with vesicles. Scale bar: 500 nm. (f) Line flp::HA820 was used to visualise FLP::HA in male and female gametocytes. FLP::HA localisation and abundance is similar in both sexes. Flp::HA820 gametocytes were PFA-fixed and stained against HA (green in females/red in males); male-specific marker (GFP) is shown in green; female-specific marker (RFP) is shown in red. Hoechst was used to visualise DNA (blue). Confocal images, scale bars: 5 μm. PFA: paraformaldehyde In order to define the expression profile and localisation of FLP during the parasite life cycle, we endogenously fused a hemagglutinin (HA) tag to the C-terminus of FLP using a PlasmoGEM vector (Gomes et al., 2015; Schwach et al., 2015). We have successfully generated an flp::HA parasite line and recycled the selection cassette (Figure S2). The FLP::HA fusion protein was detected in both schizont- and gametocyte-enriched blood stage lysates revealing the fusion protein at the expected size of 186 kDa (Figure 1c); in the haploid parasite, the tagged protein is the sole source of FLP. Antibody staining of fixed parasites confirmed expression in schizonts, showing a speckled signal (Figure 1d, Movie S1). High abundance of FLP::HA was observed in gametocytes, localising to widespread and distinct intracellular speckles (Figure 1d, Movie S2). Electron microscopy and immune labelling on thin sections of purified flp::HA gametocytes confirmed localisation to certain vesicles in the cytoplasm (Figure 1e). Immunogold typically localised close to the vesicle membranes, which is in agreement with FLP being a predicted transmembrane protein (Figure 1a,e). To investigate whether FLP::HA abundance and localisation differed between male and female gametocytes, we generated the flp::HA820 line (Figure S3). The parental 820cl1m1cl1 parasite expresses red fluorescent protein (RFP) specifically in female gametocytes, gametes, zygotes, and ookinetes, whereas GFP is present in male gametocytes and gametes (Mair et al., 2010; Ponzi et al., 2009; Sinha et al., 2014). Both male and female gametocytes showed a similar abundance and localisation of FLP::HA (Figure 1f, Movies S3 and S4). 2.2 Genetic depletion of flp in the gamete causes transmission failure P. berghei is characterised by a high genetic tractability. Attempts to delete flp using standard gene targeting approaches with the plasmid b3D.DT∧H.∧D (Janse et al., 2006) or PlasmoGEM vectors (Gomes et al., 2015; Schwach et al., 2015) however failed repeatedly in our hands, and a recent global screen identified the gene to be essential for the blood stage (Bushell et al., 2017). This strongly suggests that FLP fulfils an essential function during the intraerythrocytic development cycle. As FLP is abundant in the gametocyte stage on both protein and transcript levels (Figure 1c,d,e and Figure S4), we decided to investigate FLP function during this stage. Replacing the endogenous promoter of a gene of interest with the asexual clag (PBANKA_140060) promoter has previously allowed to maintain expression of essential blood stage genes during the intraerythrocytic cycle while reducing expression in gametocytes and gametes and thus helping to expose protein functions (Laurentino et al., 2011; Santos et al., 2015; Sebastian et al., 2012). In order to assess FLP depletion in gametocytes following promoter exchange, we generated a Δflp::HAgam line (Figure S5), in which FLP is endogenously tagged with HA and under the control of the clag promoter. FLP::HA was clearly depleted in gametocytes of the Δflp::HAgam line as shown by the lack of immuno-fluorescent signal (Figure 2). Figure 2Open in figure viewerPowerPoint Genetic depletion of FLP in gametocytes leads to a complete life cycle arrest. FLP::HA fusion protein is depleted from Δflp::HAgam gametocytes as compared with flp::HA gametocytes. PFA-fixed gametocytes were stained against HA (green); DNA was visualised using Hoechst (blue). Confocal images, scale bar: 5 μm. Fluorescent signal was quantified in z stacks using FIJI (integrated density) and plotted as mean ± SEM. Dotted line represents background fluorescence (obtained from staining of WT parasites). T test was used for statistical analysis In order to explore the role for FLP, we additionally generated an independent, HA-tag-less Δflpgam line (Figure S6). Although intraerythrocytic development of asexual blood stages and gametocytes was not affected in the Δflpgam line (Figure S7), gametocytes lacking FLP underwent a dramatic life cycle arrest (Table 1). The Δflpgam line completely failed to transmit by mosquito bite. Neither oocysts (data not shown) nor midgut and salivary gland sporozoites (Table 1) could be observed after transfer of the Δflpgam parasites to Anopheles mosquitoes by natural transmission. Consequently, the Δflpgam parasites were not transmitted to a subsequent rodent in a so-called bite back experiment and did not cause a blood stage infection in the mouse (Table 1). As expected, an independent Δflpgam clone and Δflpgam820 parasites, generated in the 820cl1m1cl1 background (Figure S8), reproduced the same transmission impairment observed in the Δflpgam line (Table 1, Figure S9a). Table 1. Δflpgam lines do not produce stages subsequent to the gametocyte stage after transmission to Anopheles mosquitoes (i.e., sporozoites) or in vitro (ookinetes) Ookinete conversion Midgut sporozoites Salivary gland sporozoites Blood stage prepatency Culture 1 Culture 2 Culture 3 Cage 1 Cage 2 Cage 3 Cage 1 Cage 2 Cage 3 Mouse 1 Mouse 2 Mouse 3 PbANKA (WT) 89% 93% 93% 14,000 10,000 18,000 5,600 9,000 14,700 3 3 3 Δflpgam 0% 0% 0% 0 0 0 0 0 0 ∞ ∞ ∞ Δflpgam clone 2 ND ND ND 0 0 0 0 0 0 ND ND ND 820cl1m1cl1 (WT) ND ND ND 10,000 3,500 8,000 2,000 2,100 2,000 ND ND ND Δflpgam820 ND ND ND 0 0 0 0 0 0 ND ND ND Note. Ookinete conversion was determined from three independent ookinete cultures. Midgut sporozoites per female mosquito were quantified 14 days after blood meal from three independent feeds. Salivary gland sporozoites per female mosquito were quantified 17–21 days after blood meal from three independent feeds. Ten mosquitoes were used for vector-to-host transmission; mice were followed by daily Giemsa smears; numbers depict blood stage positive of total mice. ND: not determined. 2.3 FLP is required for red blood cell and parasitophorous vacuole membrane rupture during gamete egress The reason for the failure of Δflpgam parasites to establish mosquito infection was identified through in vitro ookinete cultures. Unsurprisingly, induction of gametogenesis followed by 20-hr culture at 19°C with the Δflpgam or Δflpgam820 parasites yielded no ookinetes (Table 1) but instead contained large numbers of unfertilized gametocytes (Figure S9a). Ookinete development relies on the formation of mature extracellular microgametes and macrogametes from intraerythrocytic gametocytes, and fertilisation. Gamete egress is easily observed during exflagellation, a process in which flagellated motile microgametes emerge violently from the RBC (Kuehn & Pradel, 2010). Although wild type microgametes emerged freely (Figure 3a, Movie S5), microgametes of the Δflpgam line were beating vigorously within the intact RBC but were unable to egress (Figure 3a, Movie S6). Figure 3Open in figure viewerPowerPoint Life cycle arrest of Δflpgam line is caused by gamete egress failure. (a) Following gametocyte activation, the Δflpgam line fails to produce freely moving microgametes. Live cell imaging, scale bar: 5 μm. (b) Gametocytes from indicated lines were combined in ookinete cultures. Coculturing of parental wild type lines (GFPcon and 820cl1m1cl1) yielded red and green ookinetes in a ratio of 1:1. However, coculturing of the Δflpgam line with GFPcon resulted in an 83% reduction in red ookinetes production. Live cell imaging, scale bar: 5 μm. Ratios of red and green ookinetes were quantified in the bar plot. Data originate from five biological replicates. Error bars represent standard error of mean (SEM). Mann–Whitney test was used for statistical analysis. (c) Prior to activation, both PbANKA and Δflpgam gametocytes reside within the RBC membrane. WT activated microgametes (distinguished by tubulin staining in red) emerge from the host RBC. In contrast, activated microgametes of the Δflpgam line do not egress. Gametocytes were activated by xanthurenic acid and a temperature drop, PFA-fixed and stained against tubulin (red) and RBCM marker TER-119 (green). Hoechst was used to visualise DNA (blue). Confocal images, scale bar: 5 μm. (d) Prior to activation, both WT and Δflpgam gametocytes reside within the PVM. WT activated microgametes (distinguished by tubulin staining in red) emerge from the PVM. In contrast, activated microgametes of the Δflpgam line in most cases do not egress. Gametocytes were activated by xanthurenic acid and a temperature drop, PFA-fixed and stained against tubulin (red) and PVM marker SEP1 (green). Hoechst was used to visualise DNA (blue). Confocal images, scale bar: 5 μm. Graphs in (c–d) quantify in percent the different appearances of the RBCM and PVM in activated gametocytes, respectively, as analysed in 50 images in each of three independent biological replicates. Error bars represent standard error of mean (SEM). Examples of membrane scoring are shown in Figure S10. RBC: red blood cells; RBCM: red blood cell membrane; PVM: parasitophorous vacuole membrane; PFA: paraformaldehyde Whereas male egress behaviour is easily assessed by live microscopy due to the motility of the males, such defects in females are more difficult to quantify. Cross-fertilisation of a mutant with a fully fertile parasite line offers a tool to address the female fertility status (Khan et al., 2005; Mair et al., 2010). In order to investigate macrogamete egress in more detail, we used the Δflpgam820 line, generated in the 820cl1m1cl1 background introduced above, in which macrogametocytes, macrogametes, and ookinetes express RFP. To test whether Δflpgam820 macrogametes can be fertilised by fertile microgametes, we performed a cross-fertilisation experiment with the fully fertile GFPcon line, which expresses cytoplasmic GFP under the control of the constitutive eef1a promoter (Franke-Fayard et al., 2004). In vitro ookinete cultures were set up by mixing equal numbers of gametocytes from each line. As expected, crossing of the two wild type reference lines (GFPcon and 820cl1m1cl1) led to the production of red and green ookinetes in a ratio of about 1:1, attesting to the equal fertility of the two macrogamete populations (Figure 3b). In contrast, crossing of the Δflpgam820 line with GFPcon produced less than 10% red ookinetes, a significantly reduced ratio (Figure 3b). This indicates that Δflpgam820 macrogametes remain trapped within the RBC just like males, and as a consequence, their transmission competence is highly compromised. A detailed quantification by immunofluorescence assays of fixed gametocytes before and after activation revealed that unlike the wild type, Δflpgam gametocytes failed to dissolve the RBC and PV membranes. Both were still surrounding flagellated Δflpgam microgametes at time points when free microgametes were present in the wild type and both membranes were clearly disrupted (Figure 3c,d, Movies S7–S10). In most wild type parasites, the RBCM (stained for the marker protein Ter-119; Kina et al., 2000) was completely lysed, occasionally with visible remnants. All Δflpgam parasites, on the other hand, were trapped in an intact or only slightly damaged RBCM (Figure 3c). Images of the PVM staining (stained for the marker protein SEP1; Birago et al., 2003) showed that over 80% of wild type gametes were membrane free, whereas most Δflpgam gametocytes were still surrounded by the membrane, although it appeared damaged in about one third of cases (Figure 3d). Examples of membrane scoring are shown in Figure S10. We performed electron microscopy on activated gametes to shed more light on the egress failure displayed by Δflpgam parasites. While wild type gametes underwent egress after activation (with remnants of lysed PVM and RBCM often observed in the vicinity of the cell), Δflpgam gametes were surrounded by both the PVM and RBCM in all observed cases (Figure 4). Figure 4Open in figure viewerPowerPoint Transmission electron microscopy (TEM) of Δflpgam gametes reveals retention of RBCM and PVM. Gametocytes were enriched, purified, and activated by XA and drop in temperature. Electron microscopy reveals that WT gametes egress from both the RBCM and PVM, only the plasma membrane is visible (remnants of the lysed membranes can be observed in the vicinity of the cell). Δflpgam gametes are surrounded by the plasma membrane, the PVM, and the red blood cell cytosol and RBCM. Magnified views show the clear presence of only the plasma membrane (PbANKA) and the plasma membrane surrounded by PVM and RBC cytosol (Δflpgam). Arrowheads point to the three membranes. Scale bar: 500 nm. RBC: red blood cells; RBCM: red blood cell membrane; PVM: parasitophorous vacuole membrane These observations implicate FLP in the disruption of the membrane compartments surrounding the intraerythrocytic gamete. The apparent failure to dissolve the membranes explains both the in vitro ookinete formation and the in vivo transmission failures. 2.4 FLP relocalises to the cell periphery during gamete maturation RBC egress is an essential part of the process of male and female gametogenesis (Sinden, 1984; Sinden, 2015) and is mediated by the discharge of specialised secretory vesicles, including osmiophilic bodies (Koning-Ward et al., 2008; Sinden, 1982). Egress vesicles are produced in the intraerythrocytic gametocyte when still in the mammalian host and typically show a random cytoplasmic distribution. Transfer to the mosquito midgut or to a medium mimicking a similar environment results in trafficking of such vesicles to the cell surface and exocytosis of soluble proteins into the parasitophorous vacuole, leading to PVM and RBCM lysis (Figure 5a; Andreadaki et al., 2018; Deligianni et al., 2013; Kehrer, Frischknecht, & Mair, 2016; Koning-Ward et al., 2008; Olivieri et al., 2015; Ponzi et al., 2009; Sologub et al., 2011; Talman et al., 2011). Following gametocyte activation in vitro (Billker et al., 1998), FLP::HA speckles were detected at the periphery of the cell (Figure 5b, Movie S11). Relocalisation of egress vesicles was previously shown to be calcium dependent (Olivieri et al., 2015). In agreement with that, in gametocytes activated in the presence of BAPTA-AM, a membrane-permeable calcium chelator, FLP::HA-labelled speckles did not relocalise to the cell periphery (Figure 5b, Movie S12). This change in localisation is reminiscent of secretory egress vesicles that contain for example G377 (Kehrer, Singer, et al., 2016; Koning-Ward et al., 2008; Olivieri et al., 2015) or PPLP2 (Deligianni et al., 2013; Kehrer, Singer, et al., 2016; Wirth et al., 2014). Figure 5Open in figure viewerPowerPoint FLP localises to egress vesicles, and its deficiency can be rescued by detergent membrane lysis. (a) Schematic of gametocyte egress. The gametocyte contains egress vesicles (yellow and black) and is surrounded by the plasma membrane (black), parasitophorous vacuole membrane (PVM, green), and red blood cell membrane (RBCM, red). Activation leads to discharge of egress vesicles and subsequent perforation and lysis of PVM, followed by perforation and lysis of RBCM. Remnants of both membranes are usually seen in the vicinity of the egressed gamete. After gametogenesis and egress are completed, one round female gamete or eight flagellated male gametes are produced from each gametocyte; nu = nucleus. (b) FLP::HA positive vesicles localise to the cell periphery after gametocyte activation. This relocalisation can be prevented by calcium chelation. Parasites wer