ZFT is the major iron and zinc transporter in Toxoplasma gondii

Abstract

Transition metals, such as iron and zinc, are indispensable trace elements for eukaryotic life, acting as co-factors in essential processes ranging from metabolism to DNA replication. These metals can be transported into cells by an evolutionary-conserved family of metal transporters; however, how the ubiquitous mammalian parasite Toxoplasma gondii acquires essential metals has been unknown. Here, we have identified and characterised the first iron and zinc importer in T. gondii. This transporter, named ZFT, localised to the parasite plasma membrane and is essential for the parasite's life cycle. We find ZFT is regulated by iron availability and overexpression sensitises cells to excess iron and zinc. Using a conditional knockdown system, we find that knockdown of ZFT leads to reduction in mitochondrial respiration and a switch to a more quiescent lifecycle stage. To confirm transport activity, we find that knockdown of ZFT leads to a reduction in parasite-associated zinc and iron, and ZFT expression complements loss of zinc transporter activity in a yeast model. Further, expression of ZFT in Xenopus oocytes demonstrates direct uptake of iron, which is outcompeted in the presence of zinc. Overall, we have identified the first metal uptake transporter in T. gondii and demonstrated the importance of iron and zinc to the parasite. This finding advances our understanding of how this obligate intracellular parasite acquires nutrients from its host.

Keywords: Toxoplasma gondii; infectious disease; iron; metabolism; microbiology; transporter; zinc.

Figure 1.. ZFT is a ZIP-domain containing protein with dynamic localisation.

(a) Alignment of ZIP proteins from T. gondii (TGME49_261720), T. gondii (TGME49_266800), P. falciparum (PF3D7_1022300), Porospora cf. gigantea (KVP17_002880), Chromera vella (Cvel_11065), H. sapiens (HsZIP1 and HsZIP2), and Bordetella bronchiseptica (BpZIP). Key conserved residues in the binuclear metal centres (BMC) M1 and M2 in transmembrane domains a4 and a5, which have been shown to be required for metal binding, are highlighted. HK motif found in Apicomplexa and HsZIP1 and HsZIP2 highlighted. Conserved glutamate residues highlighted listed below, numbers from BbZIP sequence. (b) Alphafold model of TGME49_261720 dimer with key residues highlighted (c) Schematic of F3ΔHX ZFT-3HA_zft strain construction to endogenously tag ZFT with 3xHA epitope tags at the C-terminal under the endogenous promoter with the endogenous 3′ UTR. (d) Western blot analysis of ZFT-3HA_zft confirming the expected band size (50 kDa). Non-specific bands indicated with asterisks. CDPK1 (TGME49_301440) used as a loading control. (e) Immunofluorescence of ZFT-3HA_zft demonstrating dynamic peripheral or basal localisation. Scale bars 5 μm. (f) Percentage of ZFT-3HA_zft localisation (peripheral or basal) in respect to number of parasites per vacuole. 100 parasites were counted for each condition. Bars at mean of four independent experiments, ± SD. (g) Mean fluorescence intensity (MFI) of ZFT-3HA_zft per parasite in relation to the number of parasites per vacuoles. Each point represents individual MFI/parasite values from three independent experiments. Bars at mean of the experiments ± SD. p values from ordinary one-way ANOVA, Tukey corrected for multiple comparisons.

Figure 1—figure supplement 1.. Localisation of ZFT-HA is not dependent on the 3' UTR, but is dependent on iron availibility.

(a) Schematic of F3ΔHX ZFT-3HA_sag1 strain construction to tag ZFT with a 3xHA epitope tag at the C-terminal under the endogenous promoter with the sag1 3′ UTR. (b) Immunofluorescence of ZFT-3HA_sag1 demonstrating dynamic localisation throughout the lytic life cycle, showing peripheral and basal characterisation. Scale bars 5 μm. (c) Immunofluorescence assay of ZFT-3HA_zft 24 hr post infection in untreated and low iron (100 μM DFO) conditions. Scale bars 5 μm. (d) Immunofluorescence of ZFT-3HA_zft 24 hr post infection in untreated and 20 μM DFO conditions. (e) Quantification of ZFT-3HA_zft localisation (peripheral/basal) in untreated and 20 μM DFO conditions. Bars are the mean of three independent biological experiments, ± SD.

Figure 2.. ZFT-3HA_zft is regulated by iron, and overexpressing ZFT leads to a change in iron sensitivity.

(a) Western blot analysis showing ZFT-3HA_zft expression levels 18 hr post infection in untreated and high iron (500 μM FAC) conditions. CDPK1 was used as a loading control. (b) Quantification of ZFT-3HA_zft from three independent experiments. Bars at mean ± SD. p values from two-tailed paired t test. (c) As above, at 24 hr post infection in untreated and high iron (500 μM FAC) conditions. (d) Quantification of ZFT-3HA_zft from three independent experiments. Bars at mean ± SD. p values from two-tailed paired t test. (e) Immunofluorescence assay of ZFT-3HA_zft at 24 hr post infection in untreated and high iron (500 μM FAC) conditions. Scale bars 5 μm. (f) Quantification of ZFT-3HA_zft localisation (peripheral/basal or no signal) in untreated and high iron conditions. Bars at mean of three independent experiments, ± SD. (g) Schematic of ZFT-Ty strain construction to overexpress ZFT from the TUB8 promoter in RHtdTomato parental line. (h) Western blot of ZFT-Ty showing expected size. (i) Plaque assay of parental and ZFT-Ty parasites in untreated conditions. Quantification of plaque number (j) and area (k) for parental and ZFT-Ty parasites, p values are unpaired two-tailed t test. Points represent individual plaque areas from three independent experiments, bars at mean ± SD. p values from two-tailed unpaired t test with Welch’s correction. Graphs showing mean parasite EC₅₀ for FAC (l) and DFO (m) for RHtdTomato and ZFT-Ty parasites. Each point represents an independent experiment, bars at mean of n = 3, ± SD. p values from two-tailed unpaired t test.

Figure 2—figure supplement 1.. Characterisation of ZFT-Ty overexpression line.

(a) PCR of ZFT-Ty confirming the presence of the gene. (b) Immunofluorescence assay of parental and ZFT-Ty 6 hr post infection in untreated conditions. Scale bars 5 μm. (c) Fluorescent growth curves showing the percentage of parasite survival in the presence of excess iron. (d) Fluorescent growth curve showing the percentage of parasite survival in the presence of sodium arsenite. (e) Graphs showing mean EC₅₀ for Ars for RHtdTomato and ZFT-Ty parasites. Each point represents an independent experiment, bars at mean of n = 3, ± SD. p values from two-tailed unpaired t test. (f) Fluorescent growth curves showing the percentage of parasite survival where iron is depleted using the chelator DFO.

Figure 3.. ZFT knockdown blocks parasite replication.

(a) Schematic of F3ΔHX T7S4-ZFT under the T7S4 inducible promoter. (b) RT-qPCR showing ZFT mRNA abundance, relative to 0 hr, normalised to actin. Bars at mean of three independent experiments, ± SD. p values from two-tailed unpaired t test. (c) Schematic of F3ΔHX T7S4-ZFT-3HA_sag1 under the T7S4 inducible promoter. (d) Immunofluorescence assay showing T7S4-ZFT-3HA_sag1 expression in untreated parasites and after induction with ATc for 48 hr. Scale bar 5 μm. (e) Western blot demonstrating successful knockdown of ZFT at 24, 48, 72 and 96 hr post ATc induction. CDPK1 used as a loading control. (f) Quantification of (e), relative to 0 hr, normalised to CDPK1. Bars at mean of four independent experiments, ± SD. p values from two-way ANOVA, Tukey corrected for multiple comparisons. (g) Blue Native PAGE of T7S4-ZFT-3HA_sag1, untreated and +ATc 48 hr induction, showing a ZFT-HA complex at around 100 kDa. CDPK1 used as a loading control. (h) Plaque assay of parental and T7S4-ZFT parasites untreated and ATc treated, 6 days, showing knockdown of ZFT renders the parasite unable to form plaques in the host cell monolayer. (i) Quantification of (h) showing number of plaques, normalised to parental untreated. Bars at mean of three independent experiments, ± SD. p values from two-way ANOVA, Tukey corrected for multiple comparisons. (j) Quantification of plaques area from (h). Points represent individual plaques from three independent experiments. Bars at mean ± SD. p values from ordinary one-way ANOVA, Tukey corrected for multiple comparisons. (k) Plaque assays with addition of 200 μM FAC do not rescue ZFT knockdown parasites. (l) Quantification of plaque area generated by T7S4-ZFT parasites in untreated and high iron (200 μM FAC) conditions. Points represent individual plaque areas from three independent experiments. Bars at mean ± SD. p values are from two-tailed unpaired t test with Welch’s correction. (m) Quantification of plaque number, normalised to parental untreated. Bars at mean of three independent experiments, ± SD. p values from two-way ANOVA, Tukey corrected for multiple comparisons. (n) Quantification of plaque area generated by parental and T7S4-ZFT parasites with and without ATc, in the presence of excess iron (200 μM FAC). Points represent individual plaque areas from three independent experiments. Bars at mean ± SD. p values from ordinary one-way ANOVA, Tukey corrected for multiple comparisons.

Figure 3—figure supplement 1.. Creation and validation of the ZFT conditional knockdown line.

(a) PCR of T7S4-ZFT confirming successful integration of the DHFR-T7S4 cassette at both the 5′ and 3′ end. (b) PCR confirming successful tagging of ZFT with a 3 HA epitope tag in the T7S4-ZFT parasite line. (c) Western blot showing ZFT-HA under endogenous promoter and T7S4 promoter. (d) Quantification of above western blot, normalised to CDPK1. Bars at mean ± SD, p value from unpaired t test. (e) Uncropped BN-PAGE gels of T7S4-ZFT-3HA_sag1, untreated and +ATc 48 hr induction, showing a ZFT-HA complex at around 100 kDa. CDPK1 used as a loading control. (f) Plaque assays showing that the addition of 200 μM FAC and 0.25 μM ATc instead of the standard 0.5 μM ATc does not rescue the inability of ZFT knockdown parasites from forming plaques, 6 days. (g) Quantification of plaque assay in untreated conditions showing the number of plaques normalised to parental in untreated conditions. Bars at mean of two independent experiments, ± SD. p values from two-way ANOVA, Tukey corrected for multiple comparisons. (h) Quantification of plaque assay showing the number of plaques under high iron conditions (200 μM FAC) normalised to parental in untreated conditions. Bars at mean of two independent experiments, ± SD. p values from two-way ANOVA, Tukey corrected for multiple comparisons.

Figure 4.. ZFT knockdown reduces mitochondrial respiration and the expression of Fe–S protein SDHB.

(a) Immunofluorescence assay of T7S4-ZFT-3HA_sag1 in the absence and presence of ATc for 24 and 48 hr, staining for CPN60 (apicoplast marker). (b) Western blot analysis of T7S4-ZFT with and without ATc 48 hr staining for lipoic acid. TgPDH-E2, TgBCDH-E2 and TgOGDH-E2 were detected 48 hr post ZFT knockdown, suggesting the apicoplast is functional. (c) Immunofluorescence assay of T7S4-ZFT parasites with and without ATc staining for CPN60 (apicoplast marker) following a further round of mechanical release and invasion to examine if ZFT knockdown has a delayed apicoplast phenotype. Scale bar 5 μm. (d) Quantification of % of apicoplasts from (c) with and without ATc that appeared normal or elongated. 100 vacuoles for each condition were quantified. Bars at mean of three independent replicates, ± SD. (e) Quantification of the percentage of parasites per vacuole with and without ATc following a further round of mechanical release and invasion. 100 vacuoles for each condition were counted. Bars at mean of three independent replicates, ± SD. (f) Immunofluorescence assay as above, staining for TOM40 (mitochondrial marker). Scale bar 5 μm (g) Mitochondrial oxygen consumption rate (mOCR) of T7S4-ZFT, untreated or ATc treated for 48 hr via Seahorse assay. Quantification of basal mOCR (h) and maximal mOCR (i). Bars at mean of three independent experiments, ± SD. p values are from two-tailed unpaired t test with Welch’s correction. (j) Metabolic map demonstrating ZFT knockdown shifts parasites to a more quiescent state. Points at mean, ± SD. (k) Western blot of T7S4-ZFT SDHB-3HA, untreated and ATc treated for 48 hr. CDPK1 used as a loading control. (l) Quantification of SDHB-3HA expression, relative to uninduced, normalised to CDPK1. Bars at mean of three independent experiments, ± SD. p values are two-tailed paired t test. (m) Complex IV activity assay of T7S4-ZFT parasites with and without ATc 48 hr showing ZFT knockdown reduces complex IV activity. CDPK1 used as a loading control.

Figure 4—figure supplement 1.. Validation of lines for testing mitochondrial and apicoplast functions in ZFT knockdown.

(a) Western blot of apicoplast protein CPN60 untreated and ATc treated for 48 hr showing no accumulation of the premature (pCPN60) form. CDPK1 used as a loading control. Representative of four independent biological replicates. Quantification of TgPDH-E2 (b), TgOGDH-E2 (c), and TgBCDH-E2 (d) expression, relative to uninduced, normalised to CDPK1. Bars at mean of four independent experiments, ± SD. p values are two-tailed paired t test. (e) Quantification of basal extracellular acidification rate (ECAR). Bars at mean of three independent experiments, ± SD. p values are from two-tailed unpaired t test with Welch’s correction. (f) Schematic of F3ΔHX T7S4-ZFT SDHB-3HA strain construction to endogenously tag SDHB with 3xHA epitope tags at the C-terminal under the endogenous promoter. (g) Schematic of F3ΔHX T7S4-ZFT ABCE1-3HA strain construction to endogenously tag ABCE1 with 3xHA epitope tags at the C-terminal under the endogenous promoter. (h) PCR confirming successful integration of the 3 HA epitope tag and the CAT selection cassette into the endogenous locus in the T7S4-ZFT SDHB-3HA line. Non-specific bands marked with asterisks. (i) PCR confirming successful integration of the 3 HA epitope tag and CAT selection cassette into the endogenous locus in the T7S4-ZFT ABCE1-3HA line. Non-specific bands marked with asterisks. (j) Western blot of T7S4-ZFT ABCE1-3HA untreated and ATc treated for 48 hr. CDPK1 used as a loading control. (k) Quantification of ABCE1-3HA expression, relative to uninduced, normalised to CDPK1. Bars at mean of three independent experiments, ± SD. p values are two-tailed paired t test. (l) TEM of T7S4-ZFT parasites, untreated and ATc treated for 72 hr showing no changes in mitochondrial cristae upon ZFT knockdown. Scale bar = 2 µM and 500 nM, respectively. (m) Immuno-electron microscopy images of parental parasites in the presence of 100 μM DFO for 5 days, indicating enlarged L-vesicles in the parasites. Scale bar 500 nm.

Figure 5.. Knocking down ZFT triggers partial bradyzoite differentiation.

(a) Immunofluorescence of T7S4-ZFT parasites uninduced or induced with ATc for 48 hr. Highlighted is a region where parasites are disorganised and asynchronously replicating. Scale bar 5 μm. (b) Quantification of the percentage of parasites/vacuole in parental and T7S4-ZFT parasites in the presence and absence of ATc for 48 hr. 100 vacuoles for each condition were counted. Bars at mean of four independent replicates, ± SD. (c) TEM images of T7S4-ZFT parasites with and without ATc for 5 days. Detail highlights regions of parasitophorous vacuolar membrane (PVM). Yellow arrows point to enlarged vesicles in the PV space. Scale bar 500 nm. (d) Immunofluorescence of T7S4-ZFT-3HA_sag1 untreated and +ATc 72 hr, stained with BAG1, a bradyzoite specific marker. Scale bar 5 μm. (e) Immunofluorescence assay of T7S4-ZFT-3HA_sag1 untreated and +ATc 4 days, stained with SAG1 and DBL, a cyst wall-binding lectin. Scale bar 5 μm. Quantification of % vacuoles expressing BAG1 (f) and DBL (g). 100 vacuoles for each condition counted, bars at mean of three independent experiments, ± SD. p values from two-way ANOVA, Tukey corrected for multiple comparisons. (h) Plaque assays of parental line treated with ATc for 6 days, and T7S4-ZFT parasites, treated either with ATc for 6 days, or for 3 days following washout and 6 days recovery. (i) Quantification of plaque number recovery following ATc washout, normalised to parental parasites. Bars at the mean of four independent experiments, ± SD. p values from two-tailed unpaired t test.

Figure 6.. ZFT expression depends on the availability of zinc, and ZFT expression complements lack of zinc transport in S.cerevisiae.

(a) Plaque assays showing that the addition of 25 μM ZnSO₄ and the combination of 25 μM ZnSO₄ and 200 μM FAC cannot rescue ZFT knockdown. (b) Number of plaques after treatments. Bars at mean of n = 3, ± SD. p values from two-way ANOVA, Tukey corrected for multiple comparisons. (c) Plaque areas of parasites treated with excess zinc (25 µM ZnSO₄). Points represent individual plaque areas from two (parental -ATc) or three (all other conditions) independent experiments. Bars at mean ± SD. p values from ordinary one-way ANOVA, Tukey corrected for multiple comparisons. (d) Plaque area from T7S4-ZFT parasites in untreated and high zinc (25 μM ZnSO₄) conditions. Points represent individual plaque areas from three independent experiments. Bars at mean ± SD. p values are from two-tailed unpaired t test with Welch’s correction. (e) Graph showing mean EC₅₀ for zinc for parental and ZFT-Ty parasites. Bars at mean of n = 3, ± SD. p values from two-tailed unpaired t test. (f) Western blot analysis showing ZFT-3HA_zft expression levels 24 hr post infection in untreated, high zinc (50 μM ZnSO₄). CDPK1 used as a loading control. (g) Quantification of ZFT-3HA_zft levels at 24 hr from three independent experiments. Bars at mean ± SD. p values from one-way ANOVA, Tukey corrected for multiple comparisons. (h) Western blot after treatment with 5 uM TPEN, with additional supplementation by FAC (200 uM) or ZnSO4 (50 uM) as indicated, at 24 hr post infection. (i) Quantification of ZFT-3HA_zft levels at 24 hr from three independent experiments. Bars at mean ± SD. p values from one-way ANOVA, Tukey corrected for multiple comparisons. (j) Spot assay showing TgZFT expression rescues growth in Δzrt1/2 S. cerevisiae. Representative of two independent biological experiments. (k) Under zinc chelation (5 mM EGTA), TgZFT allows Δzrt1/2 growth. Representative of two independent biological experiments.

Figure 6—figure supplement 1.. Zinc ionophore is unable to rescue the growth phenotype upon ZFT knockdown, and no change in localisation of ZFT was observed in high iron conditions.

(a) Plaque assays showing that the addition of 2 μM Zn Ionophore does not rescue the inability of ZFT knockdown parasites from forming plaques. (b) Quantification of plaque assay showing number of plaques, normalised to parental +Zn Ionophore. Bars at mean of three independent experiments, ± SD. p values from two-way ANOVA, Tukey corrected for multiple comparisons. (c) Quantification of plaque assay showing number of plaques, normalised to parental +FAC and ZnSO₄. Bars at mean of three independent experiments, ± SD. p values from two-way ANOVA, Tukey corrected for multiple comparisons. (d) Dose–response curves showing the percentage of both parental and ZFT-Ty parasite survival in the presence of excess zinc. (e) Immunofluorescence of ZFT-3HA_zft untreated and at 24 hr treatment with 50 μM ZnSO₄ showing no change in ZFT localisation. Scale bar 5 μm.

Figure 7.. Knocking down ZFT results in a decrease in parasite-associated metal.

(a) Live cell imaging and quantification of FerroOrange stained parental parasites, untreated or treated with 200 μM FAC for 48 hr. Scale bar 5 μm. Points represent the mean fluorescence intensity (MFI) of FerroOrange/parasite from four independent experiments. Bars at mean ± SD. p value from two-tailed unpaired t test with Welch’s correction. (b) Live cell imaging and quantification of FerroOrange-stained T7S4-ZFT parasites with and without ATc for 48 hr. Points are MFI of FerroOrange/parasite from four independent experiments. Bars at mean ± SD. p value from two-tailed unpaired t test with Welch’s correction. (c) Live cell imaging and quantification of FluoZin-3AM stained parental parasites treated with 50 μM ZnSO₄ for 24 hr. Points represent MFI of FluoZin-3AM/parasite from three independent experiments. Bars at mean ± SD. p value from two-tailed unpaired t test with Welch’s correction. (d) Live cell imaging and quantification of FluoZin-3AM stained T7S4-ZFT parasites with and without ATc for 48 hr. Scale bar 5 μm. Points represent the MFI of FluoZin-3AM per parasite from three independent experiments. Bars at mean ± SD. p value from two-tailed unpaired t test with Welch’s correction. (e) Inductively coupled plasma-mass spectrometry (ICP-MS) quantification of iron/parasite (fM) from T7S4-ZFT parasites with and without ATc at 24 and 48 hr. Bars at mean (n = 7 for – ATc, n = 5 for +ATc 24 hr and n = 2 for +ATc 48 hr) ± SD. p values from one-way ANOVA, Tukey corrected for multiple comparisons. (f) ICP-MS quantification of zinc/parasite (fM) from T7S4-ZFT parasites with and without ATc 24 and 48 hr. Bars at mean (n = 5 for – ATc and +ATc 24 hr and n = 2 for +ATc 48 hr) ± SD. p values from one-way ANOVA, Tukey corrected for multiple comparisons. X-ray fluorescence microscopy (XFM) of extracellular T7S4-ZFT parasites, untreated and treated for 24 and 48 hr, showing phosphorus, sulphur, iron (g) and zinc (h). (i) XFM of untreated parasites showing iron and zinc colocalisation, along with phosphorus. Quantification of Fe (j) and Zn (k) following ZFT knockdown. Bars at mean ± SD. p values from one-way ANOVA, Tukey corrected for multiple comparisons. (l) Quantification of changes in Fe, Zn, S, and P as a percentage of untreated cells, following the addition of ATc. Bars at mean ± SD. Scale bar for all images 5 μm.

Figure 7—figure supplement 1.. Quantification of X-ray fluorescence microscopy (XFM) levels of P (mM) (a) and S (mM) (b) per parasite following ZFT knockdown at both 24 and 48 hr ATc.

Bars at mean ± SD. p values from one-way ANOVA, Tukey corrected for multiple comparisons.

Figure 8.. Expression of ZFT in Xenopus laevis oocytes mediates iron uptake.

(a) Time course for the uptake of ⁵⁵Fe²⁺ into X. laevis oocytes, uninjected or expressing ZFT. Uptake was measured in the presence of 1 mM ascorbic acid to maintain iron in the reduced state. Each data point represents the mean uptake in 10 oocytes from a single experiment, and the data is representative of 3 independent experiments, points and fitted using a first-rate equation. (b) ⁵⁵Fe²⁺ uptake at 60 mins, with or without competition with 25 µM ZnCl₂. Each bar represents the mean iron uptake of 10 oocytes from a single experiment. (c) Representative current record by two-electrode voltage clamp in oocytes expressing or not ZFT after the addition of Fe ²⁺ (added as 3 mM FeCl₃ with 1 mM ascorbic acid) and a holding potential of –30 mV. (d) Current/voltage relationship in control and ZFT expressing oocytes, in the presence and absence of 3 mM Fe²⁺. Values mean of 11 oocytes, ± SD and slopes calculated from simple linear regression.

Figure 8—figure supplement 1.. Western blot of ZFT, and raw elctrophysiology traces.

(a) Western blot of streptavidin-purified surface biotinylated proteins from Xenopus oocytes, showing a single band at the expected mass for ZFT. (b) Representative traces from oocytes injected with ZFT or uninjected in the presence or absence of Fe²⁺.

Author response image 1.. Western blot of ZFT-3HA_zft and another HA-tagged unrelated cytosolic protein, demonstrating that the lower bands are most likely nonspecific.