The Vacuolar Zinc Transporter TgZnT Protects Toxoplasma gondii from Zinc Toxicity

Toxoplasma gondii is an intracellular pathogen of human and animals. T. gondii pathogenesis is associated with its lytic cycle, which involves invasion, replication, egress out of the host cell, and invasion of a new one. T. gondii must be able to tolerate abrupt changes in the composition of the surrounding milieu as it progresses through its lytic cycle. We report the characterization of a Zn2+ transporter of T. gondii (TgZnT) that is important for parasite growth. TgZnT restored Zn2+ tolerance in yeast mutants that were unable to grow in media with high concentrations of Zn2+. We propose that TgZnT plays a role in Zn2+ homeostasis during the T. gondii lytic cycle.

proposed an important role for this organelle in controlling ionic stress during the short extracellular phase of the parasite (5,6). The PLV becomes prominent when Toxoplasma is extracellular and its proton pumps (7,8) create a proton gradient that is used for the countertransport of Ca 2ϩ (5) and other ions.
The zinc ion (Zn 2ϩ ) must be tightly regulated because both a deficiency and an excess of cytoplasmic free Zn 2ϩ are deleterious for cells (9)(10)(11). Zinc is an essential element that acts as a cofactor for a large number of enzymes and regulatory proteins and that also participates in cell signaling (12,13). More than 300 enzymes that utilize Zn 2ϩ have been identified across all enzyme classes and phyla (14). Notably, 3 to 10% of the genes encoded by the human genome, over 3,000 in total, are thought to encode proteins that interact with Zn 2ϩ , a number that is likely underestimated because new Zn 2ϩ -protein interactions are still being discovered (15)(16)(17).
Enzyme inhibition, disruption of protein folding, and induction of apoptosis are some of the proposed mechanisms by which high concentrations of Zn 2ϩ may be deleterious to cells (9)(10)(11). The consistent abundance of Zn 2ϩ in our environment during the evolution of life has introduced a selective pressure on all living organisms to evolve complex mechanisms to regulate total cellular Zn 2ϩ and intracellular free Zn 2ϩ . The total concentration of cellular zinc in eukaryotic cells typically ranges from 0.1 to 0.5 mM (18); however, most of the Zn 2ϩ in cells is bound to proteins and sequestered into so-called zincosomes (19) or lysosomal compartments. The resting intracellular free Zn 2ϩ concentration is reported to be at picomolar levels (20), and cytosolic zinc-binding proteins exhibit an affinity for Zn 2ϩ in the picomolar range (21,22). These picomolar concentrations represent less than 0.0001% of total cellular Zn 2ϩ , exemplifying the precise control of cytoplasmic free Zn 2ϩ in eukaryotic cells. Free Zn 2ϩ in the extracellular space was reported to be in the range of 5 to 25 nM in the central nervous system (23), which is more than 1,000-fold higher than the predicted intracellular concentration.
T. gondii is exposed to sharp changes of extracellular Zn 2ϩ upon egress, and we propose that the PLV plays an important role in the ability of Toxoplasma to efficiently survive these changes. In the present work, we characterized a ZnT family Zn 2ϩ transporter (TgGT1_251630, TgZnT). We localized TgZnT, characterized the phenotypic profile of conditional knockdown mutants, and used TgZnT to rescue Zn 2ϩ tolerance in a Zn 2ϩ -intolerant Saccharomyces cerevisiae yeast mutant.

RESULTS
Identification of a Zn 2؉ transporter in T. gondii. With the aim of characterizing the potential role of the PLV in the survival and thriving of Toxoplasma during its extracellular passage, an essential phase of its lytic cycle, we looked at potential transporters that localize to the PLV and that could function in the transport of ions for which a strict control is required. One of these ions, Zn 2ϩ , was especially interesting because of several reasons. First, Zn 2ϩ levels need to be tightly controlled; second, there was proteomic evidence for the presence of a zinc transporter in Toxoplasma and in a PLV-enriched fraction (ToxoDB and unpublished data); and third, evidence for the proton gradient needed for its function was demonstrated in previous work (5). The Zn 2ϩ transporter gene annotated in ToxoDB (TgGT1_251630) predicts the expression of a protein of 715 amino acids with a predicted molecular weight of 77 kDa and an isoelectric point of 5.86. We named the gene TgZnT because it is the single member of this family of Zn 2ϩ transporters annotated in the T. gondii genome. The TgZnT gene product is predicted to contain 6 transmembrane domains (Fig. 1A), forming a structure similar to that of the Escherichia coli Zn 2ϩ transporter YiiP (Fig. 1B).
We studied the phylogenetic profile of TgZnT, and for this we generated a bootstrapped neighbor-joining tree of aligned and trimmed sequences (see Fig. S1 in the supplemental material) of various ZnT family proteins from a variety of organisms as well as TgZnT and its apicomplexan orthologs (Fig. 1C). The tree analysis showed that TgZnT groups with the ZnT-2 family of plant and mammalian Zn 2ϩ transporters (24) along with orthologs in other apicomplexan parasites (including both coccidian and hemosporidian parasites) (Fig. 1C). This grouping suggests that TgZnT and its orthologs may have derived from a single gene in a distant common ancestor of plants, mammals, and apicomplexans. TgZnT also possesses the histidine and aspartic acid residues thought to be required for intramembrane Zn 2ϩ binding in transmembrane helixes II and V (Fig. 1D).
TgZnT-HA localizes to the plant-like vacuole and to cytoplasmic vesicles. To investigate the localization of TgZnT, the TgGT1_251630 gene was endogenously tagged with a 3ϫ hemagglutinin (3ϫHA) tag at its 3= end, using the ligationindependent cloning C-terminal tagging plasmid previously described (25). This approach avoids the overexpression and potential abnormal distribution of the tagged protein. Western blot analysis of a clonal parasite line expressing TgZnT-HA showed several bands around the predicted molecular weight of TgZnT plus the additional 4 kDa of the 3ϫHA tag (ϳ82 kDa) (Fig. 2B). The presence of multiple bands suggests Areas highlighted in yellow are regions used for polyclonal antibody production (Fig. 3). (B) Phyre2 modeling of TgZnT (red) shows a predicted structure similar to that of the E. coli ZnT transporter YiiP (gray). Side chains of the predicted Zn 2ϩ binding residues in transmembrane helices II and V are shown in green. (C) Unrooted tree of TgZnT, apicomplexan orthologs, and other ZnTs. The branch to which TgZnT (red, bold) and its apicomplexan orthologs (red) belongs is the ZnT2-like subfamily, which primarily transports Zn 2ϩ . (D) Multiple-sequence alignment of transmembrane (TM) helices II and V of TgZnT and its apicomplexan orthologs with various other ZnTs that transport Zn 2ϩ . The histidines and aspartic acid residues that are predicted to be part of the intramembrane Zn 2ϩ -binding site (*) are conserved in TgZnT and its apicomplexan orthologs. Cp, Cryptosporidium parvum; Pf, Plasmodium falciparum; Et, Eimeria tenella; Nc, Neospora caninum; Os, Oryza sativa; At, Arabidopsis thaliana; Hs, Homo sapiens; Sh, Stylosanthes hamata; Mm, Mus musculus; Sc, Saccharomyces cerevisiae; Bs, Bacillus subtilis; Wm, Wautersia metallidurans; Ec, Escherichia coli.
Toxoplasma Zinc Transporter that TgZnT is posttranslationally modified, which is additionally supported by the prediction of phosphorylation and methylation sites annotated in the EuPathDB (26) entry for TgZnT (TgGT1_251630) (27,28). Immunofluorescence analysis with a clonal parasite line expressing TgZnT-HA showed different distributions of the labeling in extracellular and intracellular tachyzoites ( Fig. 2C and D). In extracellular tachyzoites, TgZnT-HA localized to two prominent vacuoles, one apical and one posterior. The apical vacuole showed partial colocalization with the vacuolar-H ϩ -pyrophosphatase, a PLV marker (anti-VP1) (Fig. 2C). In intracellular tachyzoites, TgZnT-HA localized to dispersed vesicles throughout the cytoplasm which did not colocalize with the anti-VP1 labeling (Fig. 2D). We performed cryo-immuno electron microscopy (CryoIEM) of TgZnT-HA extracellular tachyzoites to obtain fine details of the TgZnT localization ( Fig. 2E to J). Gold particle labeling was observed in structures ranging from small vesicles (ϳ100 nm) (Fig. 2F) to large vacuoles (Ͼ250 nm) (Fig. 2E, G to J). Of particular note, we saw that labeling favored the invaginations into the larger vacuoles (Fig. 2Ei, H to J, arrows).
To investigate the localization of untagged, wild-type TgZnT, we generated specific antibodies against a fusion of two TgZnT loop domains (Fig. 1A, yellow) in mice. Western blot analysis of lysates from RH tachyzoites showed several bands around the expected molecular weight of 77 kDa (Fig. 3D), similar to what was observed with TgZnT-HA (Fig. 2B). We also performed immunofluorescence assays (IFAs) using polyclonal anti-TgZnT, which showed the labeling of two large vacuoles in extracellular tachyzoites, and one of them showed colocalization with the red fluorescent protein (RFP)-tagged chloroquine resistance transporter (CRT), a PLV marker (Fig. 3A) (29). In intracellular tachyzoites we observed a dispersed vesicular localization (Fig. 3B) that showing the endogenous promoter (black arrow) and the 5= UTR (light green) displaced by the DHFR selection cassette (pink) and the tet7sag4 promoter (red) with the tetracycline (Tet) repressor (yellow), followed by the coding region of TgZnT with exons (green). (D) Western blot analysis of lysates from the parental strain (the Δku80 TATi strain) and the Δznt and iΔznt mutants after growth with or without 0.5 g anhydrotetracycline (ATc). Lysate from iΔznt tachyzoites after growth in ATc did not show labeling with anti-TgZnT. Tubulin was used as a loading control.
was also seen in the IFAs of TgZnT-HA tachyzoites. These vesicles did not colocalize with the vesicles labeled by CRT-RFP.
TgZnT knockout mutants exhibit reduced growth in the presence of extracellular Zn 2؉ . To establish the role of TgZnT in the T. gondii lytic cycle, we first generated knockout mutants by inserting a dihydrofolate reductase (DHFR) resistance selection cassette at the beginning of exon 1 using the CRISPR/Cas9 system and a protospacer for this region of the gene (Fig. S2A). A Western blot analysis of lysates from a subclone (the Δznt clone) of the resulting mutants showed the absence of anti-TgZnT labeling, suggesting gene disruption (Fig. S2B). We complemented these mutants with a copy of TgZnT in an overexpression vector utilizing the tubulin promoter (pDTM3) (30). These clones (the Δznt-ZnT clones) overexpressed TgZnT, as was seen by Western blot analysis of their lysates (Fig. S2B). Immunofluorescence assays of parental strain RH and the knockout and complemented overexpressing mutants confirmed these results (Fig. S2C). The knockout mutants (the Δznt mutants) showed reduced growth in plaque assays (Fig. S2D), but the overexpression of TgZnT in the Δznt-ZnT clones was also deleterious for growth, and it was not possible to complement the growth phenotype of the Δznt mutant.
The effect of overexpression of TgZnT on parasite growth did not permit proper analysis of the specific biological functions of TgZnT, so we next created conditional mutants for TgZnT, which allowed for controlled expression of the gene. For this, we modified the endogenous TgZnT locus by inserting a tet7sag4 promoter at the 5= end of the predicted open reading frame (ORF). This element responds to anhydrotetracycline (ATc) by repressing expression of the downstream gene (Fig. 3C). Subclones (the final inducible knockdown locus [iΔznt] clones) were isolated, and Western blot analysis of lysates from these clonal lines revealed that expression was responsive to ATc (Fig. 3D).
We investigated the role of TgZnT in parasite growth, and we performed plaque assays in the presence ATc and in the absence of ATc (Fig. 4A). Plaques were significantly smaller when parasites were grown in the presence of ATc (ϩATc mutants) ( Fig. 4A and B). We next wanted to investigate if the mutants were less able to cope with high extracellular concentrations of Zn 2ϩ , and for this we first transfected mutant parasites with a red fluorescent protein and selected the cells by fluorescence-activated cell sorting. These cells allowed us to study growth by following the red fluorescence as a function of time ( Fig. 4C and D). We grew parasites (with and without ATc) in the presence of several concentrations of extracellular Zn 2ϩ (Fig. 4C and D) up to 100 M Zn 2ϩ , which did not show apparent toxicity to the host cells. Higher concentrations of Zn 2ϩ were deleterious to the host cells (not shown). The growth results showed that the parental cell line grew fine at 1 to 10 M Zn 2ϩ and that only a small decrease was observed at 25 M Zn 2ϩ . Higher concentrations of extracellular Zn 2ϩ (75 to 100 M) were deleterious to the growth of the parental cells. The ϩATc mutants were intolerant to higher concentrations of Zn 2ϩ and showed a clear and significant growth difference at 1, 10, and 25 M Zn 2ϩ . At 75 M Zn 2ϩ , the ϩATc mutants were significantly deficient in their tolerance to Zn 2ϩ . Interestingly, the zinc-dependent growth difference between the ϩATc mutants and mutants grown in the absence of ATc (ϪATc mutants) was ablated in media devoid of Zn 2ϩ supplementation (containing only contaminating Zn 2ϩ ). These results support our hypothesis that TgZnT plays a role in the extracellular Zn 2ϩ tolerance of T. gondii.
TgZnT restores Zn 2؉ tolerance to Zn 2؉ -sensitive yeast mutants. To investigate the Zn 2ϩ transport function of TgZnT, we transformed zrc1Δ::cot1Δ yeast mutants, which lack their vacuolar zinc transporters and are unable to grow in media containing high concentrations of Zn 2ϩ , with a pYES2 expression plasmid containing the cDNA for TgZnT under the control of the galactose promoter. Western blot analysis of lysates from these mutants grown in media containing galactose showed labeling with anti-TgZnT (Fig. 5A), with the mutants showing a similar multiple-band profile with bands with sizes comparable to the ones observed in Toxoplasma lysates (Fig. 3D). Plate growth assays in the presence of different concentrations of ZnSO 4 revealed that zrc1Δ::cot1Δ mutants expressing TgZnT were capable of tolerating higher concentrations of Zn 2ϩ (up to 300 M), whereas the zrc1Δ::cot1Δ mutants transfected with an empty vector tolerated only 100 M (Fig. 5B). Assays in liquid media showed that the zrc1Δ::cot1Δ mutants transfected with the empty vector pYES2 were unable to grow ( Fig. 5C and D, red line) in the presence of 100 M Zn 2ϩ , while the expression of TgZnT in the zrc1Δ::cot1Δ mutants led to a partial growth recovery ( Fig. 5C and D, blue line).

DISCUSSION
We report that the gene TgZnT, present in the T. gondii genome, encodes the only annotated, functional Zn 2ϩ transporter of the ZnT family in T. gondii and that this transporter is closely related to the ZnT2-like Zn 2ϩ transporters found in plants. In extracellular tachyzoites, TgZnT localizes to large and small vesicles that, in electron microscopy images, were shown to fuse with large vacuoles, most likely the PLV. In intracellular tachyzoites, TgZnT localizes to vesicles that did not colocalize with either VP1 or T. gondii CRT. We hypothesize that these vesicles may be acidocalcisomes, which are similar to the zincosomes described in other cell types (19,31). In this regard, there is evidence of the presence of large amounts of Zn 2ϩ in acidocalcisomes of different species, as determined by X-ray microanalysis (32), and of Zn 2ϩ transporters (ZnT) in acidocalcisomes of Trypanosoma cruzi (33) and T. brucei (34). Our laboratory previously determined the presence of Zn 2ϩ in acidocalcisomes of Toxoplasma by X-ray microanalysis of whole cells (31,35). It is likely that acidocalcisomes play a role in Zn 2ϩ Toxoplasma Zinc Transporter transport and trafficking in Toxoplasma. Preliminary data by our group suggest that a zinc transporter of another family (the ZIP family) that typically transports zinc into the cytoplasm (in the opposite direction of ZnT family transporters) localizes to these vesicles as well, lending credence to this hypothesis. The mechanism responsible for the delivery of Zn 2ϩ to the PLV or other compartments, where it would be required for the activity of metalloenzymes and other metalloproteins, has not been characterized in T. gondii or any other organism. It is possible that TgZnT distributes Zn 2ϩ to various compartments as a way of activating apo-metalloenzymes, which are inactive in low-Zn 2ϩ compartments. The distribution of Zn 2ϩ to these compartments could be a mechanism for regulating the activity of these enzymes. There are numerous metalloproteases in T. gondii that are predicted to require Zn 2ϩ as a cofactor for their activity (36)(37)(38)(39), and due to the promiscuous activity of metalloproteases, they would require tight control in order to prevent unintended proteolytic activity. The presence or absence of the required Zn 2ϩ cofactor would provide for potential regulation of their activity. The fusion of TgZnT vesicles or acidocalcisomes to the PLV in extracellular tachyzoites would also fit this model, as the PLV shares characteristics of a lysosome and proteolytic activity could be activated by the fusion of vesicles carrying Zn 2ϩ and TgZnT at their membrane. The phosphorylation sites annotated in TgZnT may play a role in the regulation of transport activity, as was described for the human ZIP7 transporter (40).
In extracellular tachyzoites, our results support the hypothesis that TgZnT plays a role in the tolerance of T. gondii to the shift to high Zn 2ϩ concentration upon egress. Our finding that the growth phenotype of the ΔTgZnT mutants was eliminated upon removal of supplementary Zn 2ϩ from the media is the most significant support for this proposed role. The ability of TgZnT to restore Zn 2ϩ tolerance when heterologously expressed in yeast mutants also provides support for its Zn 2ϩ -transporting function. Prior to egress, the dispersed nature of TgZnT vesicles in intracellular tachyzoites may allow for the rapid sequestration of Zn 2ϩ throughout the tachyzoite upon egress and a subsequent trafficking of the vesicles to the PLV for final sequestration.
In summary, this report describes a functional Zn 2ϩ transporter in T. gondii capable of rescuing Zn 2ϩ tolerance upon heterologous expression in yeast mutants. TgZnT localizes to vesicles that fuse with the PLV, and its absence in T. gondii tachyzoites causes a Zn 2ϩ concentration-dependent growth defect that becomes more pronounced with high concentrations of extracellular Zn 2ϩ . TgZnT is the first Zn 2ϩ transporter to be characterized in an apicomplexan parasite, and its existence as the sole member of this family of Zn 2ϩ transporters in these organisms suggests that its role may be conserved throughout the phylum.

MATERIALS AND METHODS
Gene identification and phylogenetic analysis. A gene (TgGT1_251630, UniProt accession number S7V0D3) annotated as a member of the solute carrier 30 family and an ortholog of ZnT-2 (UniProt accession number Q9BRI3) was cloned and sequenced. The ORF of the annotated gene in the current version of ToxoDB encodes a protein of 896 amino acids with a predicted molecular weight of 97 kDa; however, we determined through sequencing and experimental evidence that the translation initiation site annotated in a previous version of ToxoDB (TGME49_chrXII:5,501,102) was the correct one.
Generation of mutants. For C-terminal tagging of the TgZnT gene, the 3= 1,662 bp (minus the stop codon) of the gene annotated as a member of the solute carrier 30a2 family (slc30a2), TgGT1_251630, was amplified using primers P1 and P2 (see Table S2 in the supplemental material), which added the sequence required for ligation-independent cloning. The PCR product was purified using a Qiaex II gel extraction kit (Qiagen) and cloned into the pLIC-3ϫHA-CAT plasmid. The purified PCR product and plasmid were treated and combined as described by Huynh and Carruthers (25). Fifty micrograms of the sterilized plasmid pTgZnT-3ϫHA-CAT was transfected into 1 ϫ 10 7 RH Δku80 TATi parasites (41). Transfected parasites were selected with 20 M chloramphenicol, and clones were isolated by limiting dilution. The genomic DNA of the clones was isolated and screened by PCR using a primer upstream of the original amplification from TgZnT (forward primer P3) and downstream pLIC-3ϫHA-CAT reverse primer P4 (Table S2). Clones were further confirmed by Western blot analysis.
Disruption of the TgZnT gene in RH was achieved by transfecting tachyzoites with 1 g of pSAG1:: CAS9-U6::sgUPRT (catalog number 54467; Addgene) (42), with the protospacer region being replaced with a protospacer (AGGAAGGCGTTTCCCCGTCC) near the 5= end of the TgZnT coding region (modified with a New England Biolabs QuikChange site-directed mutagenesis kit by using primers P5 and P6) along with a separate dihydrofolate reductase (DHFR) drug selection cassette product generated via PCR. The parasites were selected with pyrimethamine followed by subcloning. Complementation/overexpression of TgZnT was accomplished by cloning the TgZnT gene, including the untranslated regions (UTRs) and potential promoter region, into the pCTH3 plasmid. The construct was transfected into Δznt tachyzoites and selected using chloramphenicol, followed by subcloning.
For conditional knockdown of TgZnT, primers P13 and P14 were used to introduce the protospacer CGCGTCTTCAGCTCTCGCCT into pSAG1::CAS9-U6::sgUPRT, which then became pSAG1::CAS9-U6::sgZnT. Homology regions corresponding to the region upstream of the protospacer (TTGCTCTTTCGCTTCCTCT GCTCTGCGTTCGCTG) and the region at the beginning with the translational start codon (GCGGCTTGG CTGCGCCGCCGCGCTTCTTGGAACGCGGCAT) were added to a base primer (primers P15 and P16) to amplify the promoter insertion cassette (43). Four micrograms of linearized PCR product and 1 g of pSAG1::CAS9-U6::sgZnT were transfected into RH Δku80 TATi parasites. Pyrimethamine (10 M) was added to the transfection reaction mixture 24 h later, and the population was subcloned. Anhydrotetracycline (0.5 g) was added to knock down the expression of ZnT.
Parasite cultures and generation of mutants. Tachyzoites of the T. gondii RH (25) and RH Δku80 TATi (43) strains were cultured in human telomerase reverse transcriptase (hTert) fibroblasts in Dulbecco's modified Eagle's medium (DMEM) supplemented with 1% fetal bovine serum, 1 mM sodium pyruvate, and 2 mM glutamine. Tachyzoites were obtained from infected hTert cells by passing them through a 25-gauge needle or otherwise collected from the supernatant of infected cells after natural egress. The RH Δku80 TATi strain was obtained from Boris Striepen (University of Georgia).
Plaque and growth assays. Plaque assays were performed as previously described (30) with modifications. For plaque growth assays, 125 tachyzoites were used for infection of hTert fibroblasts and allowed to grow for 10 days prior to fixing and staining. Growth assays of fluorescent cells were performed using TdTomato-expressing parasites in 96-well plates preseeded with hTert fibroblasts. Serum-free DMEM without phenol red was used for the growth assay, and ZnSO 4 and ATc were added, when appropriate, along with 4,000 tachyzoites per well. The fluorescence (594 nm) from each well was recorded every 24 h for 8 days using a SpectraMax E 2 plate spectrometer. A standard curve to determine parasite numbers was generated on the day of inoculation using known numbers of TdTomatoexpressing parasites.
TgZnT loop fusion expression and antibody production. TgZnT-LF was constructed by cloning two loops (Fig. 1A) of the TgZnT cDNA using overlapping regions. The primers used were P9 and P10 for the first part of the fusion construct and P11 and P12 for the second part. The fusion protein was cloned into the PQE80L expression vector and transformed into E. coli. After induction with 1 mM IPTG (isopropyl-␤-D-thiogalactopyranoside), soluble recombinant TgZnTLF was purified using a 1-ml HisPur nickel-affinity column (Thermo Fisher).
Antibodies against the recombinant TgZnT loop fusion protein (rTgZnT-LF) were generated in mice. Six CD-1 mice (Charles River) were inoculated intraperitoneally with 100 g of rTgZnT-LF mixed with complete Freund adjuvant, followed by two boosts with 50 g of rTgZnT-LF, with each boost being mixed with incomplete Freund adjuvant. The final serum was collected by cardiac puncture after CO 2 euthanasia. The animal protocol used was approved by the UGA Institutional Animal Care and Use Committee (IACUC).
Western blot analysis and immunofluorescence assays. Purified tachyzoites were treated with cell lysis buffer M (Sigma) and 25 units of Benzonase (Novagen) for 5 min at room temperature, followed by addition of an equal volume of 2% SDS-1 mM EDTA solution. Total protein was quantified with a NanoDrop spectrophotometer (Thermo Scientific). Samples were resolved using a 10% bisacrylamide gel in a Tris-HCl-SDS buffer system (Bio-Rad). Gels were transferred for Western blot analysis. Primary antibody dilutions were as follows: 1:100 for anti-HA (monoclonal rat; Roche) and 1:1,000 for mouse anti-TgZnT. Secondary horseradish peroxidase-labeled antibodies were used at 1:10,000 dilutions.
Immunoelectron microscopy. Extracellular T. gondii parasites endogenously expressing the C-terminal 3ϫHA tag (TgZnT-HA) were washed twice with phosphate-buffered saline (PBS) before fixation in 4% paraformaldehyde (Electron Microscopy Sciences, PA) in 0.25 M HEPES (pH 7.4) for 1 h at room temperature and then in 8% paraformaldehyde in the same buffer overnight at 4°C. Parasites were pelleted in 10% fish skin gelatin, and the gelatin-embedded pellets were infiltrated overnight with 2.3 M sucrose at 4°C and frozen in liquid nitrogen. Ultrathin cryosections were incubated in PBS and 1% fish skin gelatin containing mouse anti-HA antibody at a 1/5 dilution and then exposed to the secondary antibody, which was revealed with 10-nm protein-anti-gold conjugates. Sections were observed, and images were recorded with a Philips CM120 electron microscope (Eindhoven, the Netherlands) under 80 kV.
Yeast zinc tolerance assays. Parental and zinc-intolerant zrc1Δ::cot1Δ mutants (44) of Saccharomyces cerevisiae were transformed with the pYES2 empty vector or pYES2-TgZnT. Western blot analyses using mouse anti-TgZnT (1:1,000) were used to confirm expression in the pYES2-TgZnT-transformed cells. Yeast plate growth assays were performed on 1.5% agar plates containing a pH 6.5 complete supplement mixture lacking uracil (CSMϪUra; Sunrise Science) supplemented with 2% galactose and adjusted to various concentrations of Zn 2ϩ using ZnSO 4 . Assays were performed using 3 ϫ 10 5 yeast cells per 10-l droplet and imaged after 48 h of growth.
Liquid growth assays were performed as described by Stasic et al. (8) with modifications. Yeast cells were grown on 96-well plates in CSMϪUra with 2% galactose that was either supplemented with 100 M ZnSO 4 or not supplemented. Each well was inoculated with 6 ϫ 10 6 yeast cells in 200 l. Readings were performed every hour using a BioTek Synergy H1 hybrid tester.
Statistical analyses, modeling, alignments, and tree generation. All statistical analyses were performed using GraphPad Prism software (version 7). Modeling was performed using the Phyre2 server (45). Alignments were performed using the T-Coffee multiple-sequence alignment server (46) and manually trimmed to remove gaps. Trees were generated using the software Geneious and the Juke-Cantor algorithm and bootstrapped (100 cycles) to generate the consensus tree.