Figure legends
Figure 1: Colocalization and dynamics of egress-related vesicles during gametocyte activation. (A) Schematic of the proteins MDV1, G377, and PPLP2. SP, signal peptide; MACPF, Membrane Attack Complex/Perforin domain. (B) Vesicular localization of MDV1, G377 and PPLP2 in gametocytes. WT NF54 gametocytes were immunolabeled with rabbit anti-G377, rat anti-MDV1 and mouse anti-PPLP2 antisera (red and green) to investigate co-localization of the respective proteins. Parasite nuclei were highlighted by Hoechst 33342 nuclear stain (blue). Bar; 2 µm. DIC, differential interference contrast. Corresponding negative controls are provided in Fig. S2.(C) Quantification of protein colocalization. The Pearson´s correlation coefficient (PCC) was calculated using Fiji ImageJ2. Immunolabeling of G377 or MDV1 was defined as region of interest (n = 20). The error bars indicate mean ± SD. (D) Dynamics of G377- and PPLP2-positive vesicles during gametocyte activation. WT NF54 gametocytes were collected at 0-20 min post-activation (p.a.) and immunolabeled, using rabbit anti-G377 or mouse anti-PPLP2 antisera (green). Gametocytes were counterstained with rabbit or mouse anti-P230 antisera (red). Parasite nuclei were highlighted by Hoechst 33342 nuclear stain (blue). Bar; 2 µm. NRBS, neutral rabbit serum; NMS, neutral mouse serum. (E)Calcium-dependency of vesicle discharge following gametocyte activation. WT NF54 gametocytes were treated with 25 µM BAPTA-AM prior to activation and immunolabeled with as described in (D). Untreated gametocytes served as control. A total of 50 activated (rounded) gametocytes per setting were evaluated for the presence of the G377 or PPLP2 signal at 20 min p.a. (n = 3). Corresponding IFA images are provided in Fig. S3. The error bars indicate mean ± SD. ***p≤0.001 (One-Way ANOVA with Post-Hoc Bonferroni Multiple Comparison test; C, E). Results (B, D) are representative of three independent experiments.
Figure 2: Verification of the parasite lines to be used in BioID. (A) Localization of the bait proteins in the BioID parasite lines. Gametocytes of the PPLP2-GFP-BirA line and the G377-TurboID-GFP and MDV1-TurboID-GFP lines were immunolabeled with mouse anti-GFP antibody to highlight PPLP2, G377 and MDV1, fused to GFP and biotin ligase (green). Gametocytes were counterstained with anti-P230 antisera (red); parasite nuclei were highlighted by Hoechst 33342 nuclear stain (blue). WT NF54 gametocytes served as a control. Bar; 5 µm. (B)Protein biotinylation in the BioID parasite lines. Gametocytes of the PPLP2-GFP-BirA line and the G377-TurboID-GFP and MDV1-TurboID-GFP lines were treated with biotin (+) for 20 h (PPLP2-GFP-BirA) and 15 min (G377-, MDV-TurboID-GFP). Untreated parasite lines (-) and biotin-treated and untreated NF54 WT gametocytes served as controls. Gametocyte lysates were subjected to Western blot analysis and biotinylated proteins were detected using streptavidin-conjugated AP. Asteriks (*) highlight the bait proteins. Cloning strategy and transfection verification are provided in Fig. S4. Results (A, B) are representative of three independent experiments.
Figure 3: In silico analysis of the egress vesicle proteomes. (A) Schematic of candidate selection. Putative interactors of MDV1, G377, and PPLP2 following BioID of lines MDV1-TurboID-GFP, G377-TurboID-GFP and PPLP2-GFP-BirA were subjected to domain and functional analysis, resulting in the identification of a total of 143 egress vesicle proteins. Signal peptides (SP) were predicted using SignalP 4.1 & 5.0, transmembrane domains (TM) were predicted using DeepTMHMM, and endoplasmic reticulum (ER) retention signals were predicted using DeepLoc 2.0. Functional prediction was performed via PlasmoDB. (B) Venn diagram depicting the egress vesicle proteins grouped by bait protein. The final numbers of interactors for each parasite line are depicted in bold, the predicted sex specificity is indicated (; see PlasmoDB database). (C) Pie chart depicting the egress vesicle proteins (percentage of total numbers) according to predicted molecular function. (D) Pie chart depicting the egress vesicle proteins (percentage of total numbers) grouped by stages of peak expression (; see PlasmoDB database). RI, ring stage; TZ, trophozoite; SZ, schizont; GC II, gametocyte stage II; GC V, gametocyte stage V; OK, ookinete. Detailed information on the interactors before and after the application of selection criteria is provided in Tables S1 and S2. The corresponding GO term and sex specificity analyses are provided in Fig. S6.
Figure 4: Network analysis of the egress vesicle proteins. The 143 egress vesicle proteins were evaluated for potential interactions using the STRING database. Based on the interaction among the query proteins, different functional clusters were identified. (A) Adhesion protein/LCCL domain protein cluster with 23 proteins; (B) the RBC invasion and modification cluster with 59 proteins, including (B1) the Maurer´s clefts subcluster and (B2) the rhoptry/microneme subcluster; (C) the vesicle biogenesis cluster of 13 proteins. Detailed information on individual proteins clustering in each subnetwork is provided in Table S3.
Figure 5: Expression and localization of egress vesicle proteins in blood stage parasites. Five pSLI-HA-glmS -based parasite lines expressing select egress vesicle proteins tagged with HA were generated for expression analysis. (A) Blood stage expression of egress vesicle proteins. Lysates of asexual blood stages (ABS) and gametocytes (GC) of the respective lines were immunoblotted with rat anti-HA antibody to detect Sel1 (271 kDa), PSOP1 (53 kDa), the conserved protein PF3D7_0811600 (144 kDa), Vti1 (49 kDa) and the conserved protein PF3D7_1319900 (179 kDa). ABS and GC lysate of WT NF54 served as negative controls, while immunoblotting with rabbit antibody against the ER protein Pf39 (39 kDa) served as loading control. (B)Vesicular localization of egress vesicle proteins in gametocytes. Gametocytes of the pSLI-HA-glmS -based parasite lines were immunolabeled with rat anti-HA antibody (green) to detect the respective HA-tagged egress vesicle protein. Gametocytes were counterstained with anti-P230 antisera (red); parasite nuclei were highlighted by Hoechst 33342 nuclear stain (blue). Bar; 5 µm. Cloning strategy and transfection verification are provided in Fig. S8. Results (A, B) are representative of three independent experiments.
Figure 6: Localization and discharge of OB components. (A) Localization of PSOP1 and PF3D7_0811600 in OBs. Gametocytes of the PSOP1-HA-glmS line and PF3D7_0811600-HA-glmS line were immunolabeled with rat anti-HA antibody (green). OBs and g-exonemes were highlighted using rabbit anti-G377 or mouse anti-PPLP2 antisera (red); parasite nuclei were highlighted by Hoechst 33342 nuclear stain (blue). Samples were analyzed via Airyscan super-resolution microscopy. Bar; 2 µm. DIC, differential interference contrast. (B) Quantification of protein colocalization. The Pearson´s correlation coefficient (PCC) was calculated using Fiji ImageJ2. Immunolabeling of HA-tagged egress proteins was defined as region of interest (n = 20). The error bars indicate mean ± SD. (C) PSOP1 discharge following gametocyte activation. Gametocytes of line PSOP1-HA-glmS were collected at 0-5 min post-activation (p.a.) and immunolabeled with rat anti-HA antibody to highlight PSOP1 (green). Gametocytes were counterstained by rabbit anti-P230 antisera (red); parasite nuclei were highlighted by Hoechst 33342 nuclear stain (blue). Bar; 5 µm. DIC, differential interference contrast. Results (A, C) are representative of three independent experiments.