Fig 1.
Characteristics of the transport of Pi in Toxoplasma.
A. Kinetics of Pi uptake. Extracellular parasites were incubated in a Pi-depleted reaction medium at pH 7.4 containing 100 μM 32Pi at the indicated times, before washing by filtration and radioactivity counting. Open circles: Na+-independent uptake in the presence of 140 mM choline chloride. Filled circles: Na+-dependent uptake in the presence of 140 mM NaCl. Data are means ± SEM of 4 independent assays. *, p<0.05 (unpaired Student’s t-test at matched time of reaction). B. Ion co-transport activities. 32Pi uptake was assayed on parasites incubated in a Pi-depleted medium containing radioactive Pi as indicated in A, in the presence of NaCl, KCl, NH4Cl or LiCl compared to choline chloride (control conditions). Data are means ± SEM (n = 3 independent assays). *, p<0.05 (unpaired Student’s t-test). C. Na+ dependence for Pi uptake. 32Pi uptake was assayed on parasites incubated in a Pi-depleted medium containing radioactive Pi as indicated in A, with increasing NaCl concentrations. Medium osmolarity was maintained at 300 mOsM with choline chloride supplementation. Data are means ± SEM (n = 6 independent assays). D. Saturation curve of Pi transport. 32Pi uptake by the parasites was monitored in medium with 140 mM NaCl and various concentrations of Pi at pH 7.4, and traced with 100 μM 32Pi (25 μCi/ml assay) for 30 min. Data are means ± SEM (n = 6 independent assays). E. Competition assay for Pi transport. 32Pi uptake by the parasite was monitored in medium with 140 mM NaCl and excess non-radioactive Pi at pH 7.4, and traced with 32Pi as described in D. Data are means ± SEM (n = 3 independent assays).*, p<0.05 (unpaired Student’s t-test). F. Selectivity for Pi influx. Extracellular parasites were incubated in a Pi-depleted reaction medium at pH 7.4 containing 100 μM Pi, traced with 25 μCi of 32Pi/ml for 30 min, with added of 0.3, 3 mM SO42- or no addition (−). Data are means ± SEM (n = 4 independent assays). G. Pi uptake upon inhibition of Na+-H+-ATPase. Extracellular parasites were treated 10 min with 5 nM, 50 nM, 500 nM cipargamin or without drug (DMS0 control) prior to Pi uptake, washed and incubated 10 min in the presence of 100 μM Pi and traced with 250 μCi of 32Pi at pH 7.4, in medium containing 140 mM NaCl. Data are means ± SEM (n = 3 independent assays). *, p = 0.0013; **, p = 0.0008; ***, p<0.0001 comparing with DMSO condition (unpaired Student’s t-test). H. Pi uptake upon condition of described in the presence of 100 μM Pi and traced with 250 μCi of 32Pi at pH 7.4, in a medium containing 140 mM NaCl, plus 10 μM FCCP, 100 nM bafilomycin A1, without drug (control) or vehicle (DMSO) for 10 min. Data are means ± SEM (n = 6 independent assays). *, p<0.05 comparing with DMSO (unpaired Student’s t-test). I. 32Pi uptake was measured on parasites in a Pi-depleted medium containing 140 mM NaCl and traced with 100 μM 32Pi (25 μCi/ml assay) for 30 min in different pH ranges. pH in the incubation medium was adjusted by adding concentrated HCl (5.5, 6.0 and 6.5) or KOH (7.0, 7.5, 8.0 and 8.5). The ratio H2PO4−/HPO42− varied from 50:1 at pH 5.5 to 1:20 at pH 8.5. Data are means ± SEM (n = 4 independent assays). *, p<0.05 (One-way ANOVA using the Tukey’s test, comparing each value with that obtained at pH 5.5). J. Panel a: pH-dependence of the influx of 32Pi into extracellular parasites resuspended in medium at pH 6.4, 7.4 or 8.4. Panel b: kinetic constants at pH 6.4, 7.4 or 8.4. The Vmax and Km values were extrapolated from the 3 curves in panel a. To determine the H2PO4- and HPO42- concentration dependences, the respective concentration in the medium of these ionic forms was calculated at the Pi concentrations shown in panel a, for each pH value: 6.4, 7.4 and 8.4. Asterisks, values obtained at the three different pH values differing significantly from one another (p<0.0001; One-way ANOVA using the Tukey’s test).
Fig 2.
Motif features and expression of TgPiT in Toxoplasma.
A. Phylogenetic analysis of the PHS and PiT family proteins. Amino acid sequences from different unicellular eukaryotic species were aligned and phylogenetic analysis was performed using MEGA 5.2.2 software. The PiT family members include PHO-4 from Neurospora crassa (GenBank: AAA33607.1), ScPHO89 from Saccharomyces cerevisiae (GenBank: NP_009855.1), PfPiT from Plasmodium falciparum (GenBank: CAE30463.1), LiPHO89 p from Leishmania infantum (GenBank: XP_001466587.1) and TcPHO89 p from Trypanosoma cruzi (GenBank: XP_813912.1), The PHS family members include TcPHO84 p from T. cruzi (GenBank:XM_809326.1), LiPHO84 p from L. infantum (GenBank:AFJ96967.1), NcPHO-5 from N. crassa (GenBank: AAA74899.1), CgPHO84 from Candida glabrata (GenBank: XM_445078.1) and ScPHO84 from S. cerevisiae (GenBank: CAA89157.1). p = putative sequence based on functional motif identifications. Outgroup: HsGAPDH from Homo sapiens (GenBank: NP_002037.2). B. Conserved domains in TgPiT. TgPiT has two conserved PHO4 domains (I and II) containing GLU residues required for Pi translocation (turquoise boxes), present in the PiT of yeast (ScPHO89) and P. falciparum (PfPiT; MAL13P1.206). TgPiT contains 12 putative TMD (black boxes) and a large intracellular loop at between TMD 7 and 8. C. Expression of TgPiT. Immunoblots of Toxoplasma lysates (10E7 parasites per lane) were incubated with anti-TgPiT antibodies showing a band at ~95 kD. D. Biochemical analysis of TgPiT in isolated parasites. Solubilization of TgPiT: after washing, parasites isolated from cells were lysed in buffer containing 1% TritonX-100 (Tx-100) for 15 min before centrifugation of the lysate and collection of the supernatant (S; detergent-solubilized fraction) and pellet (P; membrane fraction) for SDS-PAGE and Western blotting using antibodies against TgPiT or the surface protein SAG1 as positive control (panel a). Surface-exposure of TgPiT: after washing, extracellular Toxoplasma were incubated 30 min in the presence of 0.1 mg/ml of proteinase K (Prot K) or reaction buffer alone at 22°C before adding PMSF to inactivate proteinase K, centrifugation to collect the supernatant (S; Proteinase K-sensitive fraction) and the pellet (P; Proteinase K-resistant fraction) for SDS-PAGE and Western blotting using antibodies against TgPiT, SAG1 (plasma membrane) and Hsp70 (cytosol) antibodies as controls for surface-proteolysis (panel a). Panel b shows the quantification of the ECL signal on immunoblots from 3 independent assays (means ± SD) and expressed in percent of ratios of S fractions to P fractions for TritonX-100 or proteinase K assays.
Fig 3.
Localization of TgPiT in Toxoplasma.
A. Panel a: Fluorescence microscopy of intracellular Toxoplasma transfected with a plasmid containing TgPiT-mCherry and visualized 16 h post-transfection. An individual z-slice is shown. Panel b: Fluorescence microscopy of extracellular Toxoplasma immunostained with rat anti-TgPiT antibodies. EF, extended focus, xy1, single z-slice. Both panels show intraparasitic puncta for TgPiT. Arrow in panel b points to a larger structure. Scale bars, 5 μm. B. Double IFA using anti-TgPiT and anti-SAG1 antibodies, showing TgPiT signal aligned along the plasma membrane (inset) or at the plasma membrane (arrows). An individual z-slice is shown. Scale bar, 5 μm. C-F. ImmunoEM of Toxoplasma-infected fibroblasts for 24 h using anti-TgPiT antibodies revealed by IgG-gold particles showing TgPiT on the VAC compartments (in C), various vesicles distributed throughout the cytoplasm (in D, arrowheads) or close to the plasma membrane (PM, in E) and at the plasma membrane (C, D, F). DG, dense granules; hc, host cell; mt, mitochondrion; P, parasite. Scale bars, 300 nm. G. IFA on extracellular Toxoplasma using anti-TgPiT and anti-SAG1 antibodies. Prior to the IFA, intracellular parasites have been incubated under normal culture conditions with 1 mM phosphate (PO4) or in phosphate-poor medium with <10 μM PO4 for 24 h before isolation. Arrows show patches of TgPiT co-localizing with SAG1 that are more pronounced under condition of low PO4 conditions. Representative images are show, from 50–65 parasites. Individual z-slices are shown. The Pearson’s correlation coefficient (PCC) was calculated based on the fluorescent signal on the whole parasite. PCC values are means ± SD; *, p<0.05 (unpaired Student’s t-test). H. Real-Time qPCR for transcriptional analyses for TgPiT at 1 mM PO4 or <10 μM PO4 at the indicated times. Data of TgPiT was normalized to parasite α-actin housekeeping gene to calculate 2–ΔΔCT values. Results are graphed as folds of induction, normalized to the TgPiT transcripts at 1 mM PO4 condition. Means ± SD, n = 4 independent assays in triplicate, no significant difference (unpaired Student’s t-test). Scale bars, 5 μm.
Fig 4.
Replication and growth of ΔTgPiT parasites.
A-B. Quantitative measurement of the replication rate of parental and ΔTgPiT parasites 24 h p.i. assessed by parasite counting per PV (A) or [3H]uracil incorporation assays (B), showing replication delay for the knockout. Data in A and B are means ± SD, n = 3 independent assays. *, p<0.05; **, p<0.01; ***, p<0.005; (unpaired Student’s t-test). C. IFA of infected HFF for 24 h with parental or ΔTgPiT parasites with anti-SAG1 antibodies showing % representative images of 40–60 PV, revealing aberrant cell shape of the knockout (arrow). Individual z-slices are shown.
Fig 5.
Growth features of ΔTgPiT in vivo.
The acute virulence of ΔTgPiT parasites was evaluated in a murine model via two routes of infections. A. 150 parasites from each strain or PBS alone were used to infect intravenously outbred mice (n = 5 mice for each strain), and the mortality (panel a) and weight (panel b) of the mice were monitored daily. Panel c shows a magnified view of each plaque assay in which monolayers of fibroblasts were infected with 150 parasites for 5 days. *, p = 0.0024; **, p<0.0001 (Log-rank Mantel-Cox test). B. 50 parasites from each strain or PBS alone were used to infect subcutaneously outbred mice (n = 6 mice for each strain) to monitor the mortality (panel a) and weight (panel b) of the mice. Note that Y axis starts at 50% for better representation of the data for each group. Plaque assays on infected fibroblasts with 50 parasites for 9 days are shown in panel c. *, p = 0.0496; **, p<0.0001 (Log-rank Mantel-Cox test).
Table 1.
Pi uptake by parental, ΔTgPiT and ΔTgPiT::PiT parasites Intracellular parasites were cultivated for 24 h in Pi-depleted DMEM with 1% FBS prior to the phosphate uptake assay, which was performed as described in legend of Fig 1, in a Pi-depleted reaction medium at pH 7.4 containing 100 μM 32Pi in the presence of 140 mM NaCl or choline chloride (ChCl) for energized and unenergized phosphate uptake conditions, respectively, at the indicated times.
Fig 6.
Determination of phosphate content in ΔTgPiT.
Extracellular parental, ΔTgPiT and ΔTgPiT::PiT parasites treated with perchloric acid were incubated with active or inactive exopolyphosphatase to measure inorganic phosphate. Total Pi corresponds to free monomeric Pi and polyP. The concentrations of polyP were deduced from values obtained on exopolyphosphatase-treated samples (giving total Pi concentrations) subtracted from values obtained on denaturated exopolyphosphatase-treated samples (giving free Pi concentrations). Values are mean ± SD, n = 3 independent assays. p values were calculated using Fisher's LSD test.
Fig 7.
Cell volume of ΔTgPiT parasites.
A. EM of HFF infected for 24 h with parental or ΔTgPiT parasites. Comparison between parental (panel a) and ΔTgPiT (panel b) parasites for PV size and parasite global morphology, showing abnormally thinner knockout parasites. Bars, 10 μm. B. Cell volume measurement of parental, ΔTgPiT and ΔTgPiT::PiT parasites using a Coulter counter. Panel a: volume distribution of parasites. Extracellular parasites were counted after gating on the basis of 6 different volumes, starting from 10 μm3 to 40 μm3, with 5 μm3 increments. The percent of parasite population was normalized to 100% of the population at the 10 μm3 gate as all parasites were found to have a larger volume that this gate. Data are mean ± SEM, n = 3 independent assays with samples in triplicate. P values calculated using Fisher's LSD test were statistically significant between parental and ΔTgPiT parasites (p = 0.0173), and between complemented and ΔTgPiT parasites (p = 0.0023). Panel b: Linear regression equations for volume assessment of parental, ΔTgPiT and ΔTgPiT::PiT parasites. Average volumes were determined based on curves of the % population decline with increased gating volumes (in panel a) from each independent biological replicate. A linear regression was calculated with Y = % of parasite population and X = parasite volume. Data are mean ± SEM (n = 3). *, p = 0.0035; **, p<0.0001 (Uncorrected Fisher's LSD).
Fig 8.
Acidocalcisomes in ΔTgPiT parasites.
A. EM of HFF infected for 24 h with parental or ΔTgPiT parasites. Comparison between parental (panel a) and ΔTgPiT (panel b) parasites for acidocalcisome (arrowheads) content. Bars, 500 nm. B. Ultrastructure of acidocalcisomes (arrowheads) in ΔTgPiT parasites, typified by luminal electron-dense inclusions. Bars, 500 nm. C-E. Panel of different acidocalcisomes in ΔTgPiT parasites, showing: in D from panel i to vii increased electron-dense material in the matrix; in E various shape and; in F proximity to other organelles. Bars in D-F, 200 nm. DG, dense granule; Go, Golgi; hc, host cell; m, mitochondrion; mi, microneme; n, nucleus; V, the VAC.
Fig 9.
Quantification of acidocalcisomes in ΔTgPiT parasites.
A. EM of representative extracellular parental and ΔTgPiT parasites incubated in normal or low PO4 medium for 24 h and applied to carbon-coated formvar grids before examination at the microscope, showing the abundance of acidocalcisomes in the mutant. B. Dotplot graphs for acidocalcisome number per parasite strain per PO4 condition (50 parasites observed in 3 independent preparations). p values were calculated using Fisher's LSD test.
Fig 10.
Development and ultrastructure of ΔTgPiT parasites upon phosphate deprivation.
A. Quantitative measurement of parental and ΔTgPiT parasite replication 24 h p.i. assessed by parasite enumeration after incubation in normal or low PO4 conditions. No statistical differences using a p-value are observed between normal and low PO4 concentrations within each parasite group. Statistical differences are observed between parental or ΔTgPiT parasites at normal and low PO4 concentrations. *, p<0.05; **, p<0.001; ***, p<0.0025 (unpaired Student’s t-test). B. Parasite growth quantification by plaque assays. Confluent monolayers of HFF were infected with 100 parental or ΔTgPiT parasites and maintained in normal or low PO4 for 7 days before counting the plaques and measuring their size, from 4 or 5 independent assays in triplicate. Panel a shows representative images of lysis plaques for the 4 conditions. Panel b are dotplot graphs for plaque number with p-values (unpaired Student’s t-test) and the table is means ± SD of plaque area (*, p<0.01 between normal and low PO4 within each parasite group). C-D. EM of HFF infected for 24 h with parental or ΔTgPiT parasites at low PO4. C. Comparison between parental (panel a) and ΔTgPiT (panel b) parasites for PV size and parasite global morphology, showing rachitic knockout parasites. Bars, 5 μm. D. ΔTgPiT parasites showing: in panel a, an abnormally enlarged dividing parasite with two nuclei (n1 and n2), two nascent apexes (arrowheads), in panel b, four misshapen parasites with one poorly dividing (arrowhead) and membranar debris (asterisks as in Fig 11A, panel b) in the PV lumen, and in panel c, 3 acidocalcisomes (arrowheads). Bars, 500 nm. DG, dense granules; hc, host cell; n, nucleus.
Fig 11.
Ultrastructure of VAC in ΔTgPiT parasites.
A. IFA on HFF infected for 24 h with parental or ΔTgPiT parasites stained with anti-CPL antibodies showing a strong signal for the mutant. Scale bars, 5 μm. B. Panel a: EM of extracellular parental or ΔTgPiT parasites collected during their egress from HFF prior to fixation, comparing VAC (arrowheads) size at the same magnification on these representative images. Bars, 500 nm. Panel b: Measurement of the VAC surface area from 18 independent electron micrographs of parasites from each group, showing a significant increase in VAC size in parasites lacking TgPiT, compared to the parental strain. Data are means ± SD. Statistical significance was determined using unpaired Student’s t-test. C. Representative EM of VAC (red arrowheads) from ΔTgPiT parasites, characterized by e-lucent content (panel a), luminal accumulation of material (panel b), or irregular shape (panel c), with the % for each phenotypes from 55 VAC. Bars, 300 nm. D. EM of a ΔTgPiT parasite showing VAC-acidocalcisome interactions. VAC, red arrowheads; acidocalcisomes, yellow arrowheads. Bars, 300 nm.
Fig 12.
Ion homeostasis in ΔTgPiT parasites.
A. Calcium regulation in ΔTgPiT parasites and the effect of NH4Cl-induced calcium release from acidic compartments. Panel a: Representative tracings of cytosolic Ca2+ measurements on extracellular parental, ΔTgPiT and ΔTgPiT::PiT parasites loaded with the ratiometric fluorescent indicator Fura-4F-AM. Each tracing is representative of 3 independent experiments. NH4Cl at 10 mM was added at 200 sec to induce Ca++ release from acidic stores. Panels b and c: Quantification of tracings from 3 independent experiments showing the ratiometric Fura-4F-AM fluorescence for the 3 parasite strains, before and after NH4Cl addition. Data are mean ± SEM. *, p = 0.0078 (KO vs. parental) and p = 0.0061 (KO vs. complemented) (Tukey's multiple comparisons test). B. pH regulation in ΔTgPiT parasites and the effect of bafilomycin A1 (BafA1). Representative tracings showing the intracellular pH of parental, ΔTgPiT and ΔTgPiT::PiT parasites measured via BCECF-AM ratiometric fluorescence. Each tracing is representative of 3 independent experiments. The V-H+-ATPase inhibitor BafA1 at 10 nM was added at 250 sec to depolarize the plasma membrane. Panel b: Quantification of tracings from 3 independent experiments showing the average intracellular pH in the three parasite strains Data are mean ± SEM. *, p<0.0001 (Tukey's multiple comparisons test).
Fig 13.
RNA-Seq comparison between ΔTgPiT and parental parasites.
A. Volcano plot revealing 281 genes having increased expression (red) and 163 with decreased expression (green), with statistical significance less than 0.05 in the ΔTgPiT relative to the parental strain. The green and red dashed lines represent the borderline of Log2 fold change of 0.4 in gene transcripts, and the genes above the black dashed line had padj values of statistical significance below 0.05. Each sample was sequenced in duplicate for statistical comparison. B-C. Left: Pie charts plotting the percent of proteins whose genes are up-regulated (in B) or down-regulated (in C), identified in the ToxoDB. Right: Pie charts categorizing the identified proteins based on their biological functions.
Fig 14.
Hypothetical model for ion homeostasis and osmoregulation in Toxoplasma.
Na+ homeostasis in Toxoplasma is accomplished via cooperation between ATP4, TgNHE1 and TgPiT. When intake of Na+ is reduced through TgPiT ablation, compensatory mechanisms may include decreased activity of ATP4 (dashed arrows) and increased activity of TgNHE1 (thick arrows), both resulting in reduced H+ concentrations in the parasite and thus cytosolic alkalinization. Consequently, the mutant decreases the export of H+ mediated by V-H+-ATPase (dashed lines). ΔTgPiT parasites have a severe constitutive loss in cytosolic ionic strength through reduced internalization of Pi and sodium into the parasite and impaired expulsion of these ions from VAC into the cytosol. Due to low Pi availability, the mutant has reduced concentrations of total Pi and polyP that is compensated for increased numbers of acidocalcisomes in an attempt to garner as much phosphate as possible.