Distinct signalling pathways control Toxoplasma egress and host-cell invasion

Abstract

Calcium signalling coordinates motility, cell invasion, and egress by apicomplexan parasites, yet the key mediators that transduce these signals remain largely unknown. One underlying assumption is that invasion into and egress from the host cell depend on highly similar systems to initiate motility. Using a chemical-genetic approach to specifically inhibit select calcium-dependent kinases (CDPKs), we instead demonstrate that these pathways are controlled by different kinases: both TgCDPK1 and TgCDPK3 were required during ionophore-induced egress, but only TgCDPK1 was required during invasion. Similarly, microneme secretion, which is necessary for motility during both invasion and egress, universally depended on TgCDPK1, but only exhibited TgCDPK3 dependence when triggered by certain stimuli. We also demonstrate that egress likely comes under a further level of control by cyclic GMP-dependent protein kinase and that its activation can induce egress and partially compensate for the inhibition of TgCDPK3. These results demonstrate that separate signalling pathways are integrated to regulate motility in response to the different signals that promote invasion or egress during infection by Toxoplasma gondii.

Figure 1

Localization of TgCDPK3 depends on putative acylation sites. Immunofluorescence analysis of parasites expressing different alleles of HA9-tagged TgCDPK3. The region within the dotted line represents an optical slice through the centre of the gliding parasite, collapsed onto the slice showing its trail. Graph represents the relative fluorescence intensity across the apical end, marked by the white arrows, of three parasites; mean±s.e.m.; red, HA9; green, SAG1.

Figure 2

Chemical genetic strategy to inhibit CDPKs in vitro and in vivo. (A) Frequency of gatekeeper residues in the 109 active kinases of T. gondii. (B) In vitro kinase activity against syntide-2. Recombinant TgCDPK1 or TgCDPK3 carrying a methionine (M) or glycine (G) gatekeeper was assayed in the presence of different concentrations of 3-MB-PP1. Means±s.e.m., n=3 experiments. (C) Genotypes of the strains used in this study at the TgCDPK1 and TgCDPK3 loci. Exons are designated by boxes, and introns by lines. The codon and amino acid for the gatekeeper residue are designated in red, as well as epitope tags introduced in the process of manipulation. (D) Immunoblot for c-myc-tagged TgCDPK1^M, or Ty-tagged TgCDPK3 following allelic replacement. Aldolase was used as a loading control.

Figure 3

TgCDPK1 and TgCDPK3 play different roles in A23187-triggered egress and host-cell invasion. (A) Invasion of fibroblasts by different strains in the presence of 5 μM 3-MB-PP1 or vehicle alone (DMSO). Extracellular and intracellular parasites were stained differentially and counted relative to the number of host-cell nuclei in each field. Student’s t-test; ***P>0.0005; means±s.e.m., n=3 experiments. (B) Egress following 20 min incubation with different concentrations of 3-MB-PP1, and treatment with 2 μM A23187 for 5 min. Egress was measured as a function of lactate dehydrogenase released from host cells, and normalized to the levels resulting from parasites treated with DMSO instead of 3-MB-PP1. Means±s.e.m., n=3 experiments. (C) A23187-induced vacuole permeabilization detected by vacuolar DsRed leakage monitored by fluorescence video microscopy of strains in the presence of 5 μM 3-MB-PP1. The time stamps represent minute:second after the addition of A23187, the circle in the first frame of each movie represents the area quantified for (D). (D) Relative fluorescence in a circular area with a diameter of 6 μM, within the vacuole. The three lines for each genotype represent measurements from three independent experiments.

Figure 4

Gliding and microneme secretion following inhibition of TgCDPK1 or TgCDPK3. (A) Types of gliding motility recorded over 4 min for each strain in the presence of 5 μM 3-MB-PP1. Student’s t-test; ***P<0.0005; **P<0.005; means±s.e.m., n=3 experiments, corresponding to 2–3 videos each. (B) Speed of gliding observed in the presence of 5 μM 3-MB-PP1. Means±s.e.m., n=3 experiments, corresponding to 2–3 videos each. (C) Effect of 5 μM 3-MB-PP1 on MIC2 secretion in response to 2% ethanol or 2 μM A23187. Student’s t-test; ***P<0.0005; **P<0.005; means±s.e.m., n=3 experiments. (D) Surface staining for MIC2 following stimulation of secretion in the presence of 5 μM 3-MB-PP1 or vehicle alone (DMSO). The ROM4 cKO was incubated with ATc where noted, and used as a control for the surface accumulation of MIC2. The histogram was generated for the intensity of surface staining of different strains in a representative FACS experiment.

Figure 5

PKG activation compensates for the inhibition of TgCDPK3 during egress. (A) Egress following 20 min incubation with different concentrations of Compound 1, and treatment with 2 μM A23187 for 5 min. Egress was measured as a function of lactate dehydrogenase released from host cells, and normalized to the levels resulting from parasites treated with DMSO instead of Compound 1. Means±s.e.m., n=3 experiments. (B) Strain CDPK3^M stimulated by 0.5 mM Zaprinast. Egress monitored by video microscopy. The time stamps represent minute:second after the addition of Zaprinast. (C) Egress following 20 min incubation with different concentrations of Compound 1, and treatment with 0.5 mM Zaprinast for 20 min. Means±s.e.m., n=3 experiments. (D) Egress triggered by different Zaprinast concentrations after treating strains for 20 min with 5 μM 3-MB-PP1. Egress was measured as described above. Means±s.e.m., n=3 experiments. (E) Comparison of egress following treatment for 20 min with or without 5 μM 3-MB-PP1, and triggered for 5 min with either 2 μM A23187 or 0.5 mM Zaprinast alone or in combination. Student’s t-test; ***P<0.0005; **P<0.005; *P<0.05; n.s., P>0.05; means±s.e.m., n=3 experiments.