The protozoan parasite Toxoplasma gondii encodes a gamut of phosphodiesterases during its lytic cycle in human cells

Abstract

Cyclic nucleotide signaling is pivotal to the asexual reproduction of Toxoplasma gondii, however little do we know about the phosphodiesterase enzymes in this widespread obligate intracellular parasite. Here, we identified 18 phosphodiesterases (TgPDE1-18) in the parasite genome, most of which form apicomplexan-specific clades and lack archetypal regulatory motifs often found in mammalian PDEs. Genomic epitope-tagging in the tachyzoite stage showed the expression of 11 phosphodiesterases with diverse subcellular distributions. Notably, TgPDE8 and TgPDE9 are located in the apical plasma membrane to regulate cAMP and cGMP signaling, as suggested by their dual-substrate catalysis and structure modeling. TgPDE9 expression can be ablated with no apparent loss of growth fitness in tachyzoites. Likewise, the redundancy in protein expression, subcellular localization and predicted substrate specificity of several other PDEs indicate significant plasticity and spatial control of cyclic nucleotide signaling during the lytic cycle. Our findings shall enable a rational dissection of signaling in tachyzoites by combinatorial mutagenesis. Moreover, the phylogenetic divergence of selected Toxoplasma PDEs from human counterparts can be exploited to develop parasite-specific inhibitors and therapeutics.

Keywords: 3′IT, 3′-insertional tagging; AC, adenylate cyclase; Apicomplexa; Bradyzoite; COS, crossover sequence; CRISPR, clustered regularly interspaced short palindromic repeats; EES, entero-epithelial stages; FPKM, fragments per kilobase of exon model per million; GC, guanylate cyclase; GMQE, Global Model Quality Estimation; HFF, human foreskin fibroblast; HXGPRT, hypoxanthine-xanthine-guanine phosphoribosyltransferase; IMC, inner membrane complex; Lytic cycle; MAEBL, merozoite adhesive erythrocytic binding ligand; MOI, multiplicity of infection; OCRE, octamer repeat; PDE, phosphodiesterase; PKA, protein kinase A; PKG, protein kinase G; PM, plasma membrane; QMEAN, Quality Model Energy Analysis; Tachyzoite; cAMP and cGMP signaling; sgRNA, single guide RNA; smHA, spaghetti monster-HA.

Graphical abstract

Fig. 1

Toxoplasma gondii harbors 18 cyclic nucleotide phosphodiesterases. Shown are the primary structure of parasite PDEs. The approximate position of PDEase I domain and other modules were predicted by PFAM, SMART and NCBI conserved domain search tools. The number and location of transmembrane helices are consensus of TMPred, TMHMM and Phobius algorithms. Images were generated with proportional scaling in IBS (v1.0.3) software. PDEase I, domain of cyclic nucleotide phosphodiesterase; GAF, a domain present in certain cGMP-specific PDEs, Adenylyl cyclases and FhlA; OCRE, Octamer repeat; MAEBL, merozoite adhesive erythrocytic binding ligand.

Fig. 2

Most Toxoplasma PDEs group with other apicomplexan phosphodiesterases and are predicted to degrade cAMP and/or cGMP. (A) A parsimonious phylogenetic tree of the PDEase domains from TgPDE1-18 with their orthologs from Homo sapiens (Hs), Drosophila melanogaster (Dm), Danio rerio (Dr), Caenorhabditis elegans (Ce), Dictyostelium discoideum (Dd), Cavenderia fasciculata (Cf), Leishmania major (Lm), Trypanosoma brucei (Tb), Eimeria tenella (Et) and Plasmodium falciparum (Pf). The cladistic analysis was performed using the Maximum Likelihood method (1000 bootstraps, gray spheres). Accession numbers (UniprotKB) along with the organism abbreviations are indicated in brackets except for human and apicomplexan PDEs (see Table S1; Tg, red; Et, blue; Pf, green). (B) Substrate specificity of PDE proteins, as predicted by their individual clustering with cAMP-, cGMP- and dual-specific consensus sequences of human PDEs (Fig. S1), or PfPDEs. The bootstrap score of individual clades (if acquired) are shown only for Toxoplasma PDEs to minimize numerical cluttering. The question-marked boxes indicate a bootstrap ≤ 50 (green or red), and the black boxes refer to outgroup clading. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 3

TgPDE1-18 are expressed in a broadly stage-specific manner during the lifecycle of T. gondii. (A) Scheme illustrating the 3′-insertional tagging (3′IT) of TgPDE1-18 genes with spaghetti monster (smHA) epitope. For each PDE, a plasmid encoding Cas9 and gene-specific guide RNA (pU6-Cas9-TgPDE1-18_sgRNA) was constructed and transfected along with the corresponding donor amplicon into the RHΔku80/hxgprt^– (parental) strain. Transgenic parasites were selected for HXGPRT expression (selection cassette, S.C.) (B) Genomic screening of the clonal strains expressing individual TgPDE1-18-smHA_3′IT proteins. Screening primers, specified as red-labeled arrows in panel A, were used to test gDNA from the transgenic (“T”) and parental (“P”, negative control) strains. (C, D) Immuno-dot and western blots of tachyzoites encoding smHA-tagged PDEs. The protein extracts were either directly loaded onto blotting membrane (C), or first resolved by 8% SDS-PAGE and then blotted (D). In both cases, blots were subjected to immunostaining for the HA-tag along with TgHsp90 (loading control). Note that the appearance of 4 extra PDEs (TgPDE5, 8, 12 and 18) in the western blot is due to 4× higher sample loading than dot blot. Visualization of the bands of expected theoretical molecular weight for TgPDE5, 8, 10, 12 and 18 requires a very high contrast and exposure, which oversaturates other PDEs including TgPDE6, 7 and 9. Hence, western blot does not allow a fair comparison of relative expression levels, which can be better assessed by dot blot. Sample loading in dot blot follows the PDEs, as numbered, and parental strain (“P”) as a negative control. (E) Heat map showing transcript expression (FPKM values) of TgPDE1-18 in tachyzoites, tissue cysts and entero-epithelial stages (EES1-EES5 [53]). EES1 = very early, EES2 = early, EES3 = mixed, EES4 = late, EES5 = very late. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 4

Phosphodiesterases expressed in tachyzoites show assorted spatial distribution. Intracellular parasite strains encoding individual smHA-tagged TgPDE1-18 proteins (TgPDE1-18-smHA_3′IT) were allowed to infect confluent HFF cells (24–36 h post-infection), and then immunostained using α-HA and α-TgGap45 antibodies. The host-cell and parasite nuclei were visualized by DAPI (scale, 2 μm). Images were acquired using clonal transgenic tachyzoites (Fig. 3). Panels with no detected smHA staining denote a lack of protein expression. In case of TgPDE5, 8 and 18, only a minor fraction of vacuoles was fluorescent that is probably due to dependence on the cell cycle and/or low transcript expression (Fig. S5A and Table 1).

Fig. 5

TgPDE8 and TgPDE9 reside in the apical plasmalemma, whereas TgPDE1, TgPDE7 and TgPDE10 are located at the periphery of tachyzoites. (A) Localization of TgPDE8 and TgPDE9 with an apical marker (TgIsp1). Intracellular parasites expressing TgPDE8-smHA_3′IT and TgPDE9-smHA_3′IT were stained with α-HA and α-TgIsp1 antibodies. (B, C) Immunostaining of smHA-tagged TgPDE8 and TgPDE9 (B), and of TgPDE1,TgPDE7 and TgPDE10 (C) in tachyzoites treated with α-toxin. Inner membrane complex and plasma membrane were stained by α-TgGap45 or α-TgSag2 antibodies, respectively, after drug-induced uncoupling of the two entities (see control assays in Fig. S5B). The merged images include DAPI-stained host and parasite nuclei in blue (scale, 2 μm). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 6

TgPDE9 exhibits a typical tertiary structure with a defined substrate pocket. (A, B) Homology models of TgPDE9 overlaid with the crystal structure of HsPDE9A (PDB, 3dyn) (A) and of HsPDE4D (PDB, 2pw3) (B). Human PDEs are depicted in gray background and corresponding models of the TgPDE9 catalytic domain are illustrated in salmon and blue, respectively. Zn²⁺ (purple) and another metal ion (e.g., Mg²⁺, green) bound to the catalytic center are also shown as spheres. (C, D) Inset view of the substrate-binding pocket of TgPDE9 with cGMP (C) and cAMP (D), as deduced by modeling based on HsPDE9A and HsPDE4D, respectively. Only key residues located in the catalytic site are indicated. (E) BIPPO-docked TgPDE9 model (salmon) superimposed with cGMP-bound HsPDE9A (gray). (F) Docking of PF-04957325 into TgPDE9 model (blue) overlaid with cAMP-bound HsPDE4D (gray). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 7

TgPDE8 and TgPDE9 can hydrolyze both cAMP as well as cGMP; and tachyzoites can survive a knockout of the latter enzyme. (A) Immunoprecipitation of the native TgPDE9-smHA_3′IT protein by means of α-HA-agarose beads. Fresh syringe-released tachyzoites (1 × 10⁹ for TgPDE8; 2 × 10⁸ for TgPDE9) were used to prepare indicated samples. Note that a much higher amount of tachyzoites was required for TgPDE8 due to its very low and heterogeneous expression (see Fig. 3C-D, S5A). TgHsp90 (red), a cytosolic protein, was used as a control. (B) Colorimetric phosphodiesterase activity assay to test the functionality and substrate specificity of TgPDE8 and TgPDE9. The enzyme assay was set up using 200 μM of substrate (cAMP or cGMP) and 6-22 μg protein. The common PDE inhibitors (BIPPO, 100 μM; IBMX, 50 μM; PF-04957325, 10 μM) were included, as indicated. The control reaction did not contain any enzyme but comprised the corresponding substrates. The substrate-free (enzyme-only) reaction was subtracted to measure the hydrolytic activity, which was normalized to the BCA-quantified protein amounts. (C) Schematics of TgPDE9 knockout by CRISPR/Cas9-assisted double homologous recombination in smHA-tagged progenitor strain (see Fig. 3A). Transgenic mutant parasites were selected by pyrimethamine. (D) Genomic PCR using crossover-specific primers confirming the events of 5′ and 3′ recombination in the Δtgpde9 mutant. (E, F) Immunofluorescence and immunoblot of tachyzoites proving a loss of apical signal in the Δtgpde9 strain. TgGap45 (immunofluorescence) and TgHsp90 (immunoblot) were visualized as control proteins. (G) Plaque assays demonstrating the growth fitness of specified strains in standard culture medium. The assay was set up by infecting confluent HFF monolayers with 150 parasites of each strain (7 days) (n=3 assays).