Analysis of CDPK1 targets identifies a trafficking adaptor complex that regulates microneme exocytosis in Toxoplasma

Abstract

Apicomplexan parasites use Ca²⁺-regulated exocytosis to secrete essential virulence factors from specialized organelles called micronemes. Ca²⁺-dependent protein kinases (CDPKs) are required for microneme exocytosis; however, the molecular events that regulate trafficking and fusion of micronemes with the plasma membrane remain unresolved. Here, we combine sub-minute resolution phosphoproteomics and bio-orthogonal labeling of kinase substrates in Toxoplasma gondii to identify 163 proteins phosphorylated in a CDPK1-dependent manner. In addition to known regulators of secretion, we identify uncharacterized targets with predicted functions across signaling, gene expression, trafficking, metabolism, and ion homeostasis. One of the CDPK1 targets is a putative HOOK activating adaptor. In other eukaryotes, HOOK homologs form the FHF complex with FTS and FHIP to activate dynein-mediated trafficking of endosomes along microtubules. We show the FHF complex is partially conserved in T. gondii, consisting of HOOK, an FTS homolog, and two parasite-specific proteins (TGGT1_306920 and TGGT1_316650). CDPK1 kinase activity and HOOK are required for the rapid apical trafficking of micronemes as parasites initiate motility. Moreover, parasites lacking HOOK or FTS display impaired microneme protein secretion, leading to a block in the invasion of host cells. Taken together, our work provides a comprehensive catalog of CDPK1 targets and reveals how vesicular trafficking has been tuned to support a parasitic lifestyle.

Figure 1.. Identifying CDPK1-dependent phosphorylation with sub-minute resolution.

(A) Stimulating parasites with zaprinast triggers Ca²⁺-mediated activation of CDPK1, resulting in the secretion of microneme proteins (red) required for motile stages of the parasite. Conditional knockdown (cKD) of CDPK1 endogenously tagged with mNeonGreen-mAID-Ty (green) after auxin treatment. (B) Flow cytometry of mNeonGreen (mNG) fluorescence in extracellular CDPK1 cKD or parental TIR1 parasites treated with vehicle or auxin for 3.5 hrs. (C) Schematic of phosphoproteomic time course. Parasites were harvested prior to CDPK1 depletion with auxin for 3.5 hrs, followed by stimulation with zaprinast or vehicle (DMSO). Samples were collected at 0, 9, 30, and 300 sec. The experiment was performed in biological replicates. Samples were labeled with TMTpro, pooled for analysis, and phosphopeptides were enriched using SMOAC prior to LC-MS/MS. Four sets of samples were generated: enriched phosphoproteomes for zaprinast [1] and DMSO [3], and proteomes for zaprinast [2] and DMSO [4]. Mock reporter ion intensities enabling relative quantification for a given peptide are shown to illustrate fold-change of unique phosphopeptide abundances during zaprinast stimulation. CDPK1-dependent phosphorylation is determined by calculating the area under the curve (AUC) difference between vehicle and auxin treatment conditions. (D) Protein abundances in the zaprinast proteome set [2] at 300 sec comparing vehicle- and auxin-treated CDPK1 cKD parasites. (E) UpSet plot for the number of phosphopeptides identified in the enriched zaprinast phosphoproteome [1] across individual replicates. Phosphopeptides exhibiting CDPK1-dependent phosphorylation with p < 0.05 are indicated. (F) Scatter plot of AUC_difference values of enriched zaprinast phosphopeptides [1] across biological replicates. Significance was determined by comparing the distribution of AUC_difference values from zaprinast phosphopeptides [1] to a null distribution of DMSO phosphopeptides [3] for individual replicates. (G) CDPK1-dependent and zaprinast-dependent phosphopeptide abundances over time. Ratios of zaprinast-treated samples relative to the vehicle-treated (no auxin) t = 0 samples. Median ratios of a group (solid lines). Individual phosphopeptides (opaque lines). CDPK1-dependent phosphopeptides (Group A) determined as described in F. Zaprinast-dependent phosphopeptides (Groups B, C, and D) were determined by comparing the distribution of AUC_vehicle values from zaprinast phosphopeptides [1] to a null distribution of DMSO phosphopeptides [3]. Groups were determined by projection-based clustering. (H) GO terms enriched among phosphopeptides undergoing a significant change after zaprinast stimulation. Significance was determined using a hypergeometric test.

Figure 2.. Myristoylation modulates CDPK1 activity and alters its interacting partners.

(A) Complementation strategy used to evaluate the functional importance of CDPK1 myristoylation. See Figure 2—figure supplement 1 for the construction of the iKD line. (B) Immunoblot demonstrating the auxin-dependent depletion of endogenous CDPK1 in the iKD, cWT, and cMut parasites (Myc) as well as equivalent expression of the complements (HA). T. gondii (toxo) antibody was used as a loading control. (C) Biochemical validation of complemented lines by YnMyr-dependent pull down. Enrichment of WT and Mut complements (HA). The inducible endogenous CDPK1 (Myc) and T. gondii (toxo) antibody was used as enrichment and loading controls, respectively. (D) Localization of the complemented versions of CDPK1 and corresponding cytosolic reporters within cWT (GFP) and cMut (mCherry) by immunofluorescence. (E) Myristoylation-dependent subcellular partitioning of CDPK1. Localization of YnMyr-enriched CDPK1 was evaluated using differential centrifugation. The partitioning into pellet [P] and supernatant [S] fractions was detected by immunoblot (CDPK1) and compared to doubly acylated GAP45. GFP and SAG1 were used as S and P controls, respectively. As only half of the supernatant fraction was removed from the high-speed pellet (100,000 × g), the GFP signal is present in the latter. (F) Partitioning of complemented WT and mutant CDPK1 after high speed centrifugation (HA). T. gondii (toxo) antibody was used as a P control whereas GFP and mCherry were used as S controls for cWT and cMut, respectively. (G) Plaque assays demonstrating that myristoylation of CDPK1 is important for the lytic cycle of T. gondii. (H) Lack of CDPK1 myristoylation delays ionophore-induced egress from host cells. Each data point is an average of n = 3 biological replicates, error bars represent standard deviation. Significance calculated using 1-way ANOVA with Tukey’s multiple comparison test. See Figure 2—figure supplement 1 for vehicle controls. (I) Immunoprecipitation-MS (IP-MS) of CDPK1-HA in cWT, cMut, and untagged TIR1 parasites across n = 2 biological replicates. Significantly enriched proteins with at least one unique peptide are highlighted based on the following thresholds: significant enrichment in both cWT and cMut (orange) with p_cWT < 0.05, p_cMut< 0.05, cWT log₂ fold-change > 1, cMut log₂ fold-change > 1, significant enrichment in exclusively cWT (blue) or cMut (red) with similar criteria across both pull downs; t-tests wer Benjamini-Hochberg corrected. (J) Fold-enrichment comparing cWT and cMut pull-downs of significantly enriched proteins from I; t-tests were Benjamini-Hochberg corrected.

Figure 3.. Identifying the direct substrates of CDPK1.

(A) Schematic describing a strategy to identify direct substrates of CDPK1. WT (CDPK1^G) and mutant (G128M; CDPK1^M) parasites were grown in SILAC media for multiplexed quantitation. Extracellular parasites were semi-permeabilized with aerolysin, enabling diffusion of small molecules but not proteins. CDPK1 substrate labeling was initiated by treating semi-permeabilized parasites with Ca²⁺, KTPγS, ATP, and 1B7. While CDPK1 in both WT and mutant parasites can utilize ATP to phosphorylate substrates, only WT parasites can use KTPγS to thiophosphorylate substrates. Thiophosphorylated peptides were specifically enriched and the remaining flow-through was saved for whole proteome analysis. Enriched and whole proteome samples were analyzed by LC-MS/MS. (B) 1B7 nanobody treatment inhibits non-specific extracellular kinase activity of CDPK1. Thiophosphorylated substrates were detected in lysates using an anti-thiophosphate ester antibody immunoblot. Extracellular CDPK1 activity (lane 1) was blocked by 1B7 (lane 2). Aerolysin treatment resulted in intracellular labeling (lane 5) that was unaffected by 1B7 (lane 6). (C) Thiophosphorylation performed in aerolysin-treated parasites comparing WT (CDPK1^G) and mutant (CDPK1^M) strains. Detection was performed as in B. Tubulin was used as a loading control. (D) Heatmap quantification of peptides using LC-MS/MS. Fold-change of peptide abundance shown as a ratio of WT (CDPK1^G) to mutant (CDPK1^M) abundances. Experiment was performed in n = 3 biological replicates. (E) Abundances of unique peptides after thiophosphorylation in CDPK1^G and CDPK1^M parasites across n = 3 biological replicates. Significantly enriched phosphorylated peptides are colored in red (-log₁₀(p)*fold-change > 4), one-tailed t-test. (F) GO terms enriched among significant phosphopeptides from E. Significance was determined using a hypergeometric test. (G) Putative targets of CDPK1 determined by sub-minute phosphoproteomics and thiophosphorylation of direct substrates. For a given CDPK1 target gene, the presence of a unique peptide phosphorylated in a CDPK1-dependent manner (column 1) is indicated if identified in the time course (green) and/or thiophosphorylation (magenta). The presence of additional unique phosphorylated peptides exhibiting zaprinast-dependent effects (column 2) is indicated if the peptide was phosphorylated (orange) or dephosphorylated (blue). Numbered boxes indicate multiple unique peptides. Fitness scores (column 3) obtained from genome-wide KO screen data (blues). Lower scores indicate gene is required for lytic stages of the parasite. Gene names (left), TGGT1 gene IDs (right). Gene names with asterisks (*) are associated with published data. (H) Signaling diagram describing parasite motility. Proteins exhibiting CDPK1-dependent phosphorylation by either sub-minute phosphoproteomics or thiophosphorylation are indicated (green). Proteins exhibiting CDPK1-independent phosphorylation (red) or dephosphorylation (blue) are indicated.

Figure 4.. HOOK is required for host cell invasion, but dispensable for egress.

(A) Schematic of T. gondii and H. sapiens HOOK protein domains. HOOK domain (blue), coiled-coil domain (yellow), sites phosphorylated by CDPK1 (red). (B) Immunoblot of HOOK conditional knockdown parasites (AID-HOOK) after auxin treatment for 40 hrs compared to untagged TIR1 parasites. CDPK1 was used as a loading control. (C) AID-HOOK is visualized in fixed intracellular parasites by immunofluorescence after auxin treatment for 24 hrs. Hoechst and MIC2 are used as counterstains. (D) Plaque assays of host cells infected with TIR1 or AID-HOOK parasites for 8 days in auxin. Host cells are stained with crystal violet. (E) Replication assays of host cells infected with TIR1 or AID-HOOK parasites in auxin for 24 hrs. Parasites per vacuole were quantified from immunofluorescence on fixed intracellular parasites. p > 0.9. Two-way ANOVA. (F) Invasion assays of untagged TIR1, CDPK1-AID, and AID-HOOK parasites treated auxin for 40 hrs. Medians are plotted for n = 3 biological replicates (different shades of gray); n.s., p > 0.05, Welch’s t-test. (G) Parasite egress stimulated with zaprinast following treatment with auxin for 24 hrs. Egress was monitored by live microscopy. Percent egress plotted for n = 3 biological replicates, n.s., p > 0.05, Welch’s t-test. (H) HOOK tagged with a C-terminal 3xHA in CDPK1 cKD parasites (CDPK1-AID) visualized in fixed intracellular parasites by immunofluorescence as in D.

Figure 5.. The HOOK complex is required for microneme exocytosis.

(A) Reciprocal IP-MS of HOOK-3xHA and FTS-3xHA. FTS is tagged with a C-terminal 3xHA epitope at the endogenous locus (FTS-3xHA). IP enrichment is shown as the fold-change of protein abundances in tagged versus untagged strains determined by LC-MS/MS across n = 3 biological replicates. Significantly enriched proteins with more than 3 unique peptides highlighted (red); p_HOOK < 0.05, and p_FTS < 0.05; ANOVA was Benjamini-Hochberg corrected. (B) FTS-3xHA visualized in fixed intracellular parasites by immunofluorescence after treatment with auxin for 24 hrs. Hoechst and MIC2 are used as counterstains. (C) Reciprocal IP-MS of HOOK-3xHA and 306920-3xHA. TGGT1_306920 is tagged with a C-terminal 3xHA epitope at the endogenous locus (306920-3xHA). IP enrichment is shown as a fold-change of protein abundances in tagged versus untagged strains determined by LC-MS/MS across n = 3 and n = 2 biological replicates for the HOOK-3xHA and 306920-3xHA IP, respectively. Significantly enriched proteins with more than 3 unique peptides highlighted (red); p_HOOK < 0.05, and p₃₀₆₉₂₀ < 0.05; ANOVA was Benjamini-Hochberg corrected. (D) Immunoblot of FTS cKD parasites. FTS is tagged with an C-terminal mAID-HA at its endogenous locus (FTS-AID) and treated with auxin for 40 hrs. ALD is used as a loading control. (E) Plaque assays of host cells infected with TIR1 or FTS-AID parasites for 8 days in auxin. Host cells are stained with crystal violet. (F) Micronemes are visualized in fixed intracellular FTS-AID and TIR1 parasites by immunofluorescence after treatment auxin for 24 hrs. Hoechst and GAP45 are used as counterstains. (G) Invasion assays of untagged TIR1, AID-HOOK, and FTS-AID parasites treated auxin for 40 hrs. Medians are plotted for n = 3 biological replicates (different shades of gray), n.s., p > 0.05, Welch’s t-test. (H) Parasite egress stimulated zaprinast following auxin treatment for 24 hrs. Egress was monitored by live microscopy. Percent egress plotted for n = 3 biological replicates, n.s., p > 0.05, Welch’s t-test. (I) Proximity labeling MS of FTS using TurboID (FTS-TurboID) compared to a cytosolic TurboID control (cytosolic mNeonGreen-TurboID). Protein abundances determined by LC-MS/MS are shown for n = 3 biological replicates. Significantly enriched proteins in FTS-TurboID are colored in red (red and blue), unique peptides > 3, ratio > 1, p < 0.05, ANOVA was Benjamini-Hochberg corrected. (J) Microneme protein secretion assays of parasites treated with auxin for 40 hrs. Extracellular parasites are stimulated with 1% ethanol (EtOH) and 3% IFS for 1.5 hrs. Percent MIC2 secreted is plotted for n = 3 biological replicates, n.s., p > 0.05, Welch’s t-test.

Figure 6.. CDPK1 activity and HOOK are required for microneme trafficking during parasite motility stages.

(A) Schematic to analyze microneme trafficking during parasite motile stages. Intracellular parasites expressing microneme protein CLAMP endogenously tagged with mNeonGreen (CLAMP-mNG). Parasites are treated either with 3-MB-PP1 (inhibit CDPK1) or auxin (for conditional knockdown). Live microscopy was performed to detect CLAMP-mNG signal over time. Zaprinast was added at 1 min or 30 sec to stimulate microneme relocalization to the apical end of the parasite. Fluorescence intensities across the apical-basal axis of each individual parasite within a vacuole was measured across time. Microneme relocalization was quantified by calculating the difference of maximum CLAMP intensity between time points preceding drug addition and egress. (B) Maximum intensity projections at single time points of CLAMP-mNG parasites treated with 3 μM 3-MB-PP1 or vehicle and zaprinast. (C) Relative fluorescence intensity of CLAMP-mNG signal across the apical-basal axis of parasites in B. Zaprinast (red) or vehicle (blue). Splines mean intensity for all parasites in each vacuole are shown with SD shaded. (D) Microneme relocalization. SuperPlots showing vacuole median peak differences are displayed as triangles. Individual parasites are displayed as circles. Replicates are differentially shaded, n.s., p > 0.05, unpaired t-test. (E) Maximum intensity projections at single time points of TIR1/CLAMP-mNG parasites treated with auxin and stimulated with zaprinast. (F) Microneme relocalization was quantified for TIR1/CLAMP-mNG parasites as in D. (G) Maximum intensity projections at single time points of AID-HOOK/CLAMP-mNG parasites treated auxin and stimulated with zaprinast. (H) Microneme relocalization was quantified for AID-HOOK/CLAMP-mNG parasites as in D. (I) Maximum intensity projections of extracellular TIR1/CLAMP-mNG and AID-HOOK/CLAMP-mNG parasites. (J) Percent of extracellular parasites in I with WT CLAMP-mNG localization, n.s., p > 0.05, Welch’s t-test. (K) Percent total CLAMP-mNG signal intensity in the apical versus body of extracellular parasites, n.s., p > 0.05, Welch’s t-test.

Figure 7.. Ultrastructure expansion microscopy reveals HOOK is required for apical microneme positioning.

(A) Maximum intensity projection of fixed extracellular parasites subjected to expansion microscopy. Parasites were pretreated with auxin. Acetylated tubulin (Lys40) is used to visualize cortical microtubules (green) and MIC2 to visualize micronemes (magenta). (B) 3D reconstruction of maximum intensity projections. Filaments are constructed for cortical microtubules (green) and globular organelles constructed for micronemes (magenta). (C) Percent of all micronemes localized to the apical region (4 μm from parasite apex) in each parasite. Mean ± s.d. plotted for n = 14–18 parasites, n.s., p > 0.05, Welch’s t-test. (D) Median shortest distance between individual micronemes and closest cortical microtubule per parasite. Median with 95% confidence interval plotted for n = 14–18 parasites, n.s., p > 0.05, Welch’s t-test. Actual distances in expanded samples were used. The expansion factor was 4X.