TgIF2K-B Is an eIF2α Kinase in Toxoplasma gondii That Responds to Oxidative Stress and Optimizes Pathogenicity

Toxoplasma gondii is a single-celled parasite that infects nucleated cells of warm-blooded vertebrates, including one-third of the human population. The parasites are not cleared by the immune response and persist in the host by converting into a latent tissue cyst form.

IMPORTANCE Toxoplasma gondii is a single-celled parasite that infects nucleated cells of warm-blooded vertebrates, including one-third of the human population. The parasites are not cleared by the immune response and persist in the host by converting into a latent tissue cyst form. Development of tissue cysts can be triggered by cellular stresses, which activate a family of TgIF2 kinases to phosphorylate the eukaryotic translation initiation factor TgIF2a. Here, we establish that the TgIF2 kinase TgIF2K-B is activated by oxidative stress and is critical for maintaining oxidative balance in the parasite. Depletion of TgIF2K-B alters gene expression, leading to accelerated growth and a diminished ability to convert into tissue cysts. This study establishes that TgIF2K-B is essential for the parasite's oxidative stress response and its ability to persist in the host as a latent infection. KEYWORDS Apicomplexa, Toxoplasma, differentiation, oxidative stress, parasites, stress response, translation, translational control T oxoplasma gondii is a protozoan parasite that infects nucleated cells of warmblooded vertebrates, including up to one-third of the human population (1). Toxoplasma is an obligate intracellular pathogen that replicates inside a parasitophorous vacuole (tachyzoites) that forms an interface between the parasites and their host cell (2). One of the features that facilitated Toxoplasma's omnipresence in the animal kingdom is the ability to persist in its host as latent tissue cysts (bradyzoites) that serve to disseminate the parasite to new hosts via predation (3). Toxoplasma has thus evolved mechanisms to balance parasite metabolism, replication, and cyst formation to allow rapid dissemination and encystment throughout host tissues without killing the host (4).
In addition to facilitating transmission, tissue cysts afford Toxoplasma protection from immune defenses, thus preventing the parasite from being eliminated from the host. However, a consequence of Toxoplasma persistence as a cyst is diminished parasite replication. Rapidly proliferating tachyzoites respond to environmental stresses by slowing parasite growth and by triggering differentiation into bradyzoites (3). A critical inverse relationship between the rate of cell proliferation and resistance to environmental stresses has been noted among other unicellular eukaryotes (5). This balance provides Toxoplasma with resistance to physiological and environmental stresses and ensures optimal infection and transmission into new hosts.
An important mechanism for mitigating stress damage in eukaryotes is the integrated stress response (ISR). The ISR features a family of related protein kinases that phosphorylate the a subunit of eukaryotic initiation factor 2 (eIF2a) in response to a variety of stresses. Phosphorylation of eIF2a damps global translation, which conserves nutrients and energy while promoting preferential expression of genes involved in stress remediation (6). Phosphorylation of Toxoplasma eIF2a (TgIF2a-P) occurs in response to environmental stresses and is a contributing factor in the formation of tissue cysts (7,8). Four TgIF2a kinases have been identified in the parasite, designated A through D (9). TgIF2K-A is localized to the parasite endoplasmic reticulum (ER) and responds to perturbations in this organelle, processes analogous to those of the eIF2a kinase PERK (EIFAK3/PEK) (10). TgIF2K-C and -D are related the eIF2a kinase GCN2 (EIFAK4) and respond to amino acid starvation and stresses of the extracellular environment, respectively (11)(12)(13). The function of TgIF2K-B remains unknown.
In this study, we addressed the functions of TgIF2K-B in Toxoplasma replication, stress responses, and pathogenesis. As Toxoplasma is an aerobic parasite that needs to limit molecular damage caused by generation of excessive reactive oxygen species (14), including those generated by host defense mechanisms (15), we hypothesized that TgIF2K-B functions in antioxidation responses. We generated a genetic knockout for use in in vitro and in vivo host model systems to show that TgIF2K-B is essential for the activation of catalase antioxidative responses, which affects tachyzoite replication and bradyzoite conversion. Our results suggest that control of antioxidant responses is central to the parasite's growth rate, which is a critical determinant in the ability of the parasite to persist in its host in a latent form. We conclude that TgIF2K-B is a novel sensor protein governing the rate of Toxoplasma replication and cyst formation through translational control mechanisms involving TgIF2a phosphorylation.

RESULTS
Knockout of TgIF2K-B accelerates Toxoplasma replication. We showed previously that TgIF2K-B is localized to the parasite cytosol and can phosphorylate TgIF2a in vitro (8). To address stress activation of TgIF2K-B and the role of this eIF2a kinase in parasite physiology and differentiation, we generated knockout clones in the type II ME49 strain using CRISPR (clustered regularly interspaced short palindromic repeats)/ Cas9 methods (16). The knockout strategy involved genetic insertion of a modified allele of the gene encoding dihydrofolate reductase (DHFR), which confers resistance to pyrimethamine, into the first exon of the TgIF2K-B locus (17) (Fig. 1A). PCR primers upstream of the insertion cassette (P1) were used to amplify the genomic sequences and primers inside the DHFR gene and in the TgIF2K-B gene were used to amplify across the DHFR insertion (P2). The PCR products indicated the anticipated DHFR insertion, generating ME49 Dtgif2k-b parasites (Fig. 1B). In addition, we confirmed the loss of TgIF2K-B mRNA and protein in the Dtgif2k-b parasites by RT-qPCR (Fig. 1C) and by immunoblotting using a specific antibody (8) (Fig. 1D).
It was noted that parasite clones lacking TgIF2K-B appeared to grow faster than the wild type (WT), which was confirmed by counting the number of parasites per vacuole in infected human foreskin fibroblasts (HFFs). We counted parasites at 24 and 36 h postinfection, finding that Dtgif2k-b parasites outpaced WT at each time point (Fig. 1E and F). These results suggest that the TgIF2K-B pathway plays a role in the regulation of parasite replication. Attempts to complement Dtgif2k-b parasites were problematic due to the length of the TgIF2K-B gene (8.4 kb); however, a second independent TgIF2K-B gene knockout clone phenocopied the original, supporting the idea that the reported effects are due to the disruption of the TgIF2K-B locus (data not shown).
TgIF2K-B participates in the oxidative stress response. To address the function of TgIF2K-B during infection and assess which stress condition(s) activates TgIF2K-B, we treated WT and Dtgif2k-b parasites with the following stress agents: thapsigargin (TG), an ER stress inducer that leads to activation of TgIF2K-A (10), halofuginone (HF), which inhibits the aminoacylation of tRNA Pro and is also a potent inducer of the eIF2a kinase GCN2, and sodium arsenite (Ars), which triggers general oxidative stress. Extracellular parasites were used to ensure that the observed effects were due to direct consequences of the stress on the parasites and not indirect effects through the host cell. We detected increased levels of TgIF2a-P in WT parasites exposed to each of the  Fig. 2A). In contrast, Dtgif2k-b parasites displayed minimal levels of TgIF2a-P upon treatment with arsenite (Ars), whereas the other stress conditions showed robust TgIF2a-P ( Fig. 2A). As expected after phosphorylation of TgIF2a, global protein synthesis was lowered by arsenite stress in WT parasites; however, it remained unaffected in Dtgif2k-b parasites exposed to arsenite ( Fig. 2B; also, see Fig. S1 in the supplemental material). Also of note is the higher level of basal protein synthesis in Dtgif2k-b parasites compared to WT, consistent with the demands of faster replication (Fig. 2B). In support of the idea that oxidative stress induced by arsenite is critical for induction of TgIF2a-P, cotreatment of the parasites with arsenite and the antioxidant N-acetylcysteine (NAC) led to a sharp decrease in TgIF2a-P levels in the WT parasites (Fig. 2C). We conclude that TgIF2K-B is an eIF2a kinase activated by oxidative stress in Toxoplasma.
To address the transcriptome changes resulting from TgIF2K-B deletion, we performed transcriptome sequencing (RNA-seq) analyses of WT and Dtgif2k-b parasites. Differentially expressed genes in Dtgif2k-b parasites were selected using log 2 fold change of #61 or $61 and P values of #0.05 as detailed in Materials and Methods. Loss of TgIF2K-B resulted in induction of 1,162 genes and lowered expression of 1,686 genes relative to WT parasites ( Fig. 3A and B, Table S2). The RNA-seq analyses showed that loss of TgIF2K-B led to enhanced levels of Toxoplasma superoxide dismutase-1 (TgSOD1) mRNA and downregulation of genes related to antioxidation, including PRX2 and thioredoxin (Fig. 3A). These results suggest that TgIF2K-B directs expression of genes involved in mitigating oxidative damage.
It is noteworthy that loss of TgIF2K-B resulted in upregulation of 311 genes involved in protein synthesis, including 117 ribosome-related genes, whereas 6 ribosomerelated genes were downregulated (Fig. 3B). These results further support the idea that there are higher levels of protein synthesis in Dtgif2k-b parasites that would contribute to increased replication ( Fig. 1E and F and 2B). Furthermore, the absence of TgIF2K-B , 50 nM halofuginone (HF), or 5 mM sodium arsenite (Ars) for 2 h at 37°C. Parasites were harvested, and the levels of phosphorylated TgIF2a (TgIF2a-P) and total TgIF2a were measured by immunoblotting. (B) WT and Dtgif2k-b tachyzoites were exposed to 5 mM Ars or vehicle for 2 h and then treated with puromycin for 15 min. Lysates were then analyzed by immunoblotting with puromycin-specific antibody. Levels of translation were quantified using densitometry and normalized to total TgIF2a for each respective lane. Levels of translation are presented in a bar graph (means and SD; n = 3). **, P , 0.01; ns = not significant. (C) WT parasites were treated with 5 mM Ars for 2 h in the presence or absence of 20 mM Nacetylcysteine (Nac); TgIF2a-P and total TgIF2a were then assessed by immunoblotting. To more fully address the redox functions of TgIF2K-B, we measured reactive oxygen species (ROS) levels in WT and knockout parasites, using monensin as a positive control. Results indicate that deletion of TgIF2K-B leads to significantly higher levels of ROS than in WT parasites (Fig. 4A). It has been suggested that ROS act as second messengers regulating the balance between cell proliferation and cell cycle (18,19). An early step in antioxidation responses involves SOD, which generates H 2 O 2 that will subsequently be decomposed by catalase into oxygen and water. We first measured TgSOD1 mRNA levels by reverse transcriptase quantitative PCR (RT-qPCR), confirming that they are enhanced in Dtgif2k-b parasites, as detected in the RNA-seq ( Fig. 3A and 4B). We next measured expression of catalase (TgCAT) mRNA in WT and Dtgif2k-b parasites after treatment with sodium arsenite for 2 h. While WT parasites exhibit increased TgCAT mRNA levels upon arsenite treatment, Dtgif2k-b parasites fail to do so (Fig. 4C). Consistent with the changes in mRNA expression, we found that the activity of SOD is higher in Dtgif2k-b parasites than WT parasites (Fig. 4D). Furthermore, we detected an increased concentration of H 2 O 2 in parasites lacking TgIF2K-B (Fig. 4E). Together, these results support the idea that TgIF2K-B activation and phosphorylation of TgIF2a alter expression of antioxidant genes.
To address the transcriptome changes upon arsenite treatment and determine the contributions of TgIF2K-B to this gene expression network, we treated extracellular WT and Dtgif2k-b tachyzoites for 2 h with 5 mM arsenite and then performed RNA-seq analyses. The analyses indicated that the expression of 780 genes was enhanced in WT parasites in response to arsenite treatment and 452 genes were downregulated ( Fig. 5A to C, Table S3). A major cell process enhanced in WT parasites in response to arsenite involves oxidation-reduction (Fig. 5D, Table S5), including induction of catalase, as TgIF2K-B Manages Toxoplasma Oxidative Stress Response ® noted above (Fig. 4C). In contrast, a majority of genes downregulated by arsenite treatment in WT parasites are involved in metabolism, including genes related to acetyl coenzyme A (acetyl-CoA) carboxylase, glycosyl transferase, beta-ketoacyl-acyl carrier protein synthase, beta-ketoacyl-acyl carrier protein synthase, and amino acid transporters ( Fig. 5E and Table S4). We also compared the RNA-seq analyses between WT and Dtgif2k-b parasites in response to arsenite treatment. In sharp contrast to the hundreds of genes altered in the WT response, Dtgif2k-b parasites upregulated only 15 genes and downregulated 16 genes ( Fig. 5B and C). WT and Dtgif2k-b parasites shared only 2 genes that were upregulated in response to arsenite (Fig. 5B). The results emphasize that the loss of TgIF2K-B greatly dysregulates the parasite's response to oxidative stress.
Loss of TgIF2K-B leads to parasite resistance during macrophage infection. We next investigated how the compromised ability to respond to oxidative stress might affect parasites in macrophages, which produce ROS as a defense mechanism against Toxoplasma (20, 21). We addressed this by analyzing replication of WT and Dtgif2k-b parasites in activated macrophages. Prior to infection, we stimulated J774.1 macrophages with lipopolysaccharide (LPS) for 24 h; we then quantified the number of intracellular parasites 4 days postinfection. Unlike HFFs, parasites inside macrophages are difficult to count, so they were quantified using the PCR-based B1 assay (22). While the number of WT parasites was significantly decreased in activated macrophages, replication of Dtgif2k-b parasites was unimpeded (Fig. 6A). We also found that Dtgif2k-b parasites showed diminished TgIF2a phosphorylation compared to WT parasites following infection of activated macrophages (Fig. S4). These results suggest that TgIF2K-B is important for sensing oxidative stress generated by host immune cells, and the inability to do so allows tachyzoite replication to proceed unfettered.
In addition, we measured parasite viability following infection of macrophages that had been activated and those that had not. Mock-or LPS-stimulated J774.1 macrophages were infected with WT or Dtgif2k-b parasites for 2 days, at which point parasites were harvested and used to infect an HFF cell monolayer (Fig. 6B). After 6 days, the area of parasite plaques in the HFF monolayer was measured for each condition. Results show that cycling through macrophages significantly reduced HFF infectivity of WT tachyzoites, but the Dtgif2k-b parasites displayed resistance to this insult (Fig. 6B). The effect was more pronounced when the parasites were cycled through activated macrophages. These findings further support the idea that TgIF2K-B senses oxidative stresses and triggers a response that slows parasite growth to manage the stress.
TgIF2K-B is a sensor protein governing replication rate and cyst formation. As phosphorylation of TgIF2a accompanies bradyzoite differentiation and tissue cyst formation (8), we hypothesized that parasites lacking TgIF2K-B may show deficits in tissue cyst formation. WT and Dtgif2k-b parasites were allowed to spontaneously differentiate in HFF cells. After 5 days, Dtgif2k-b cultures contained ,5% tissue cysts whereas WT cultures contained ;50% (Fig. 7A). Vacuoles containing Dtgif2k-b parasites were twice the size of those produced by WT parasites; moreover, clear Dolichos biflorus lectin staining indicative of a cyst wall was rarely evident (Fig. 7B and C). We also performed an analysis of tissue cyst formation following alkaline stress, which is a potent trigger WT parasites formed robust tissue cysts at a frequency of ;90%, whereas only ;10% of Dtgif2k-b parasites showed evidence of cyst wall formation (Fig. 7D). These results suggest that TgIF2K-B is a critical sensor that balances developmental transitions between replication and latency.
To address the importance of TgIF2K-B in vivo, we infected BALB/c mice with increasing doses of WT or Dtgif2k-b tachyzoites. The lack of TgIF2K-B had a profound effect on pathogenesis in the mouse model: BALB/c mice tolerate a dose of 10 4 WT parasites, but none of the mice were able to survive the equivalent dose of Dtgif2k-b tachyzoites (Fig. 7E). These results mirror the augmented replication rate of Dtgif2k-b tachyzoites seen in vitro and further underscore the relevance of TgIF2K-B as a sensor of in vivo signals that trigger conversion to latent forms.

DISCUSSION
The ability of Toxoplasma to persist in its host as latent tissue cysts is a key feature that has greatly enhanced the transmission of the parasite. Tissue cysts remain infectious and allow the parasite to infect new hosts via predation. Conversion of tachyzoites to bradyzoites can occur spontaneously in certain cell backgrounds (23) or following exposure to a wide variety of cellular stresses (3). We previously linked the phosphorylation of parasite eIF2a with stress-induced bradyzoite development and latency stages in Toxoplasma and its fellow apicomplexan parasite Plasmodium (9). Four eIF2a kinases have been identified in Toxoplasma that respond to ER and nutrient stresses, but the function of TgIF2K-B, which has no clear counterpart in other eukaryotes, had yet to be resolved. Here, we report the discovery that TgIF2K-B contributes to the management of the parasite's response to oxidative stress.
ROS are a natural consequence of cellular metabolism and have been reported to regulate cellular growth rate. Cells must have systems in place to sense ROS and adapt accordingly. As arsenicals induce ROS that generate oxidative stress, we examined the response of Toxoplasma to sodium arsenite. Exposure of the parasites to arsenite induced translational control, as evidenced by TgIF2a phosphorylation and diminished protein synthesis (Fig. 2). Using a CRISPR/Cas9 approach, we successfully generated a genetic knockout of TgIF2K-B, which failed to phosphorylate TgIF2a in response to oxidative stress. The ability to initiate translational control in response to ER and nutritive stresses remained unaffected in Dtgif2k-b parasites.
Dtgif2k-b parasites exhibited a higher replication rate than WT parasites. We performed an RNA-seq analysis under normal culture conditions and found abnormal transcriptome changes in Dtgif2k-b parasites that featured gene networks involved in protein synthesis and cellular metabolism, consistent with the enhanced parasite replication that was also observed (Fig. 1E and 3). We postulated that the accelerated growth may stem from increased levels of ROS, which have been reported to serve as secondary messengers that influence replication rate in other species (18). Notably, Dtgif2k-b parasites showed significant increases in ROS in the absence of added stress (Fig. 4A). The precise association between ROS levels and proliferation rate is an unresolved question for future investigation.
Abnormal alterations in gene expression occurred between WT and Dtgif2k-b parasites in response to arsenite treatment as well. RNA-seq analysis revealed that loss of TgIF2K-B resulted in dysregulation of oxidation-reduction processes upon arsenite exposure (Fig. 5). The data suggest that without TgIF2K-B, parasites fail to sense oxidative stress and consequently do not slow growth to adapt to this response. Such a phenotype would not be expected to be viable over time; indeed, Dtgif2k-b parasites have been observed to revert to normal growth rates after prolonged culture through an unresolved mechanism (data not shown). Together, these results indicate that TgIF2K-B is central for the management of oxidative stress in Toxoplasma, a process that is linked to parasite replication.
It is curious that with loss of translational control and sharply altered mRNA expression, Dtgif2k-b parasites showed increased replication in infected HFF (Fig. 1E and F) and in activated macrophages (Fig. 6). There was also enhanced morbidity in mice infected with Dtgif2k-b parasites compared with those infected with WT parasites, consistent with the faster replication of the former (Fig. 7E). Our RNA-seq analysis provides mechanistic insights into why tachyzoite replication is enhanced and cyst formation diminished in Dtgif2k-b parasites. For example, AP2IX-9 is a transcriptional repressor that restricts development of the tissue cyst; its increased expression in Dtgif2k-b parasites would inhibit the expression of bradyzoite mRNAs. Similarly, TgBFD1 is a master regulator that drives bradyzoite differentiation (24); its decreased expression in Dtgif2k-b parasites would be expected to impair tissue cyst formation. Bradyzoite-specific genes like that for LDH2 were also decreased in Dtgif2k-b parasites relative to WT parasites. Consistent with these observations, the loss of TgIF2K-B significantly lowered the frequency of tissue cyst conversion in vitro; moreover, the few cyst-like structures that did form were larger and contained more parasites (Fig. 7). In combination with the observed higher replication rates of Dtgif2k-b parasites, these results suggest that loss of TgIF2K-B dysregulates the transcriptional program essential for triggering parasite differentiation.
These results indicate that TgIF2K-B contributes to growth plasticity that enables Toxoplasma to appropriately lower the replication rate in response to oxidative stress. To achieve stress adaptation, there is an inverse correlation in unicellular eukaryotes between growth rate and stress survival (5). Diminished growth rates during stress are critical for appropriate modulation of gene expression and redistribution of resources to best mediate stress tolerance. In the case of Toxoplasma, a critical mode of stress adaptation involves differentiation into bradyzoite cysts, a clinically relevant process that is compromised in Dtgif2k-b parasites.
The phenotypes associated with loss of TgIF2K-B are representative of antagonistic pleiotropy (25), where the gene regulates multiple traits. Enhanced replication is a beneficial trait associated with Dtgif2k-b parasites, but it comes at the expense of the parasite being able to modulate proliferation when faced with signals to transition into latent stages. Therefore, TgIF2K-B functions as a critical sensor to balance parasite replication and latency; parasites lacking TgIF2K-B would likely have reduced transmission through the predation route due to their compromised ability to form infectious tissue cysts.
Disruption of TgIF2K-B genomic locus. The Dtgif2k-b parasites were generated by disrupting the genomic locus using CRISPR/Cas9 and an associated guide targeting the sequence ACCTCTCGCTCT GCGTTCCCT to allow integration of a minigene encoding a modified dihydrofolate reductase (DHFR) that confers pyrimethamine resistance (16,17). The pSAG1:U6-Cas9:sgRNA-TgIF2K-B vector was generated using Q5 site-directed mutagenesis (New England Biolabs) (26) and the listed forward and reverse primers (Table S1). After transfection, the parasites were selected using 1 mM pyrimethamine for three passages before cloning by limited dilution in 96-well plates. Independent clones were selected and assayed by PCR using primer pairs indicated in Fig. 1A.
Immunoblotting for TgIF2K-B. Parasites were purified from HFFs using a 3-mm filter, and lysates were made by suspension of the parasite pellets in a phosphate-buffered saline (PBS) solution supplemented with 0.01% Triton X-100. Proteins in parasite lysates were resolved with NuPAGE using 4 to 12% gradient bis-Tris polyacrylamide gels (Thermo Fisher Scientific). Separated proteins were then transferred from the gels to a nitrocellulose membrane. Immunoblotting was performed using antibody specific to TgIF2K-B (8) diluted 1:10 in blocking solution (Tris-buffered saline [TBS], pH 7.4, with Tween 20 and 2.5% bovine serum albumin [BSA]). The secondary antibody, goat anti-rabbit IgG (heavy plus light chain [H1L]) conjugated to horseradish peroxidase (HRP), was used at a dilution of 1:1,000. After washing, the membrane was incubated with Pierce enhanced chemiluminescence (ECL) immunoblot substrate to visualize the proteins levels using the FluorChem E system (Protein Simple).
Measurements of mRNA levels. RNA was isolated from parasites using TRIzol LS (Invitrogen), and cDNA was generated using Omniscript (Qiagen). RT-qPCR was carried out using specific primers (Table S1) with SYBR green real-time PCR master mixes (Invitrogen) and the StepOnePlus Real system (Applied Biosystems). Relative levels of transcripts were calculated with the DDC T method using the GAPDH gene as an internal control. Each experiment was performed three times, each with three technical replicates.
Measurements of TgIF2a phosphorylation. Tachyzoites were purified from infected HFFs by syringe passage and filtration as described in reference 27 and then incubated in DMEM supplemented TgIF2K-B Manages Toxoplasma Oxidative Stress Response ® with 5% FBS in the presence of a stress agent or vehicle: 1 mM thapsigargin (TG; Sigma-Aldrich), 50 nM halofuginone (HF; Sigma-Aldrich), or 5 mM sodium arsenite (Ars; Sigma-Aldrich), for 2 h at 37°C. For some reactions, 20 mM N-acetylcysteine (NAC; Sigma-Aldrich) was added. Parasites were lysed in PBS containing 0.01% Triton X-100, supplemented with the protease inhibitor cocktail cOmplete and an EDTA-free protease inhibitor cocktail (Roche). Total protein levels were quantified using a Pierce bicinchoninic acid (BCA) protein assay kit (Thermo-Fisher). Proteins were separated by NuPAGE on 4 to 12% bis-Tris gels (Thermo Fisher Scientific) and transferred to a nitrocellulose membrane for immunoblot analyses using antibodies recognizing total TgIF2a or phosphorylated TgIF2a (1:1,000) (27). The secondary antibodies, goat anti-rabbit IgG (H1L) conjugated to HRP, were used at a 1:5,000 dilution. After washing, the membranes were incubated with Pierce ECL substrate to visualize the proteins.
Parasite replication assays. Tachyzoites were allowed to invade HFF monolayers for 2 h. Infected cultures were then washed and incubated with fresh culture medium. At 24 and 36 h postinfection, infected HFF monolayers were fixed with 2.5% paraformaldehyde for 20 min and then blocked with PBS supplemented with 2% BSA. Cells were incubated with anti-SAG1 (Invitrogen) in blocking buffer containing 0.2% Triton X-100 for 1 h. Next, goat anti-mouse Alexa-Fluor 488 (Invitrogen) was applied for 1 h, followed by ProLong Gold antifade mountant with 49,6-diamidino-2-phenylindole (DAPI) to visualize parasites. The number of parasites in 250 randomly selected vacuoles was counted for each sample. All samples were anonymized prior to parasite counting.
J774.1 macrophages (ATCC) were incubated with 10 ng/ml of LPS for 24 h before being infected with tachyzoites for 2 h at an MOI of 3 (28). Following 4 days of infection, cells were harvested and the numbers of parasites were quantified using the PCR-based B1 assay (22). To test parasite fitness in HFFs after cycling through macrophages, infected macrophages were harvested by scraping and syringe lysis at 2 days postinfection. The same volume of medium containing tachyzoites was then used to infect an HFF monolayer for 2 h, which was then washed with DMEM twice. Six days postinfection, the degree of host cell lysis was determined by crystal violet staining as described previously (29).
Puromycin incorporation. To determine the levels of protein synthesis in WT and Dtgif2k-b parasites, extracellular tachyzoites were exposed to 5 mM arsenite for 2 h. Parasites were then incubated with 10 mg/ml puromycin (Sigma) for 15 min, and parasites were lysed in PBS containing 0.01% Triton X-100. Total protein synthesis was measured by immunoblotting using the antipuromycin antibody at 1:500 (EMD Millipore). Protein synthesis was quantified by densitometry using ImageJ on each lane and normalized by immunoblot using antibody specific to steady-state levels of TgIF2a protein (13).
RNA preparation and sequencing. Tachyzoites were allowed to invade HFF monolayers for 2 h; then, infected monolayers were washed, and the medium was replaced with fresh medium. At 36 h postinfection, intracellular tachyzoites were harvested from HFFs by syringe passage and filtering. Parasites were then washed in DMEM and centrifuged at 300 Â g for 15 min. Purified extracellular WT and Dtgif2k-b tachyzoites were treated with vehicle or 5 mM sodium arsenite for 2 h at 37°C, and total RNA was extracted using RNeasy (Qiagen). The RNA concentration for each sample was measured using Nanodrop One (Thermo Scientific). Library preparation and sequencing was performed by GENEWIZ (South Plainfield, NJ) using an Illumina HiSeq instrument with 150-bp paired-end reads. Three biological replicates were included in each of the untreated test groups, and four biological replicates were included in each treated test group. The quality of the raw sequencing reads obtained from GENEWIZ was checked using FastQC (30), and the low-quality reads were trimmed using TrimGalore (31). Trimmed reads were aligned to the reference T. gondii ME49 genome (https://toxodb.org/toxo/app) using hierarchical indexing for spliced alignment of transcripts (HISAT2) with default parameters (32). We obtained 335,646,715 read pairs in total, of which 269,152,478 mapped uniquely to the Toxoplasma genome (80.2%). Trimming, quality filtering, and alignment were all performed using the high-throughput computing cluster Karst and the Data Capacitor II high-performance file system, both of which are available to all Indiana University researchers. The numbers of reads aligning to the exons of different genes were counted using featureCounts (33). Last, differential expression analysis was carried out using DESeq2 with default parameters except that genes with total counts less than 10 across all samples were discarded (34). Volcano plots were created using R (35), and genes with a log 2 fold change of $61 and P value of #0.05 were considered significant for further network analyses. Gene Ontology enrichment analyses were performed using the collated genes that were significantly upregulated or downregulated, with a focus on molecular and biological cellular processes.
Reactive oxygen species. ROS were measured using the OxiSelect in vitro ROS/RNS assay kit (Cell Biolabs, Inc.) per the manufacturer's instructions. Briefly, lysates from 10 7 parasites were added to a 96well plate, followed by 50 ml of the catalyst reagent in the kit. Reaction mixtures were incubated for 5 min at room temperature before the addition of 29,79-dichlorofluorescin (DCFH) solution into each well for 1 h incubation. Fluorescence was measured at 480 nm/530 nm (excitation/emission wavelengths) in a Synergy H1 microplate reader (BioTek). All samples were assayed in triplicate. Parasites were treated with monensin (MO; 1 ng/ml) for 30 min as a positive control.
Superoxide dismutase activity and H 2 O 2 measurement. ME49 WT and Dtgif2k-b tachyzoites were purified from infected HFFs by syringe passage and filtration. Parasites were then lysed in ice cold Tris-HCl, pH 7.4, containing 0.02% Triton X-100 and supplemented with cOmplete protease inhibitor cocktail (Roche), and the lysate was clarified by centrifugation at 14,000 Â g for 5 min at 4°C. The supernatant was then used to measure superoxide dismutase (SOD) activity using a superoxide dismutase activity assay kit (Colorimetric) following the manufacturer's instructions (Abcam). The activity of SOD was measured using this colorimetric method as the optical density at 450 (OD 450 ) and compared to a standard curve.
To measure H 2 O 2 levels, purified parasites were resuspended in buffer (150 mM NaCl, 50 mM Tris-Cl, pH 7.4) supplemented with cOmplete protease inhibitor cocktail (Roche) and sonicated. For deproteinization, 4 M perchloric acid (PCA) was added to a final concentration of 1 M in the homogenate solution, followed by protein precipitation and centrifugation. The supernatant was then used in a hydrogen peroxide assay kit as per the manufacturer's instructions (Abcam). The reaction mixture was incubated at room temperature for 10 min; then each well was measured at excitation/emission wavelengths of 535/587 nm. The values of each well were compared to a standard curve to determine H 2 O 2 concentration.
Bradyzoite differentiation assays. ME49 tachyzoites were allowed to infect HFF monolayers for 2 h. After washing and replacing the medium with DMEM-5% FBS, infected cells were cultured at 37°C or subjected to the bradyzoite induction triggers alkaline stress and CO 2 deprivation. To visualize tissue cyst walls, infected cells were fixed with 2.5% paraformaldehyde and stained with rhodamine-conjugated Dolichos biflorus agglutinin (Vector Laboratories). Tissue cysts were counted in 250 randomly chosen infected cells and considered positive if there was a clear indication of D. biflorus agglutinin staining at the former PVM (parasitophorous vacuole membrane), indicating cyst wall formation. Tissue cyst area was measured using ImageJ by drawing a region of interest (ROI). All samples were anonymized prior to counting.
Analysis of infection in mice. Five-to six-week-old female BALB/c mice were purchased from The Jackson Laboratory (Bar Harbor, ME) and allowed to acclimate in our facility for 1 week. Following acclimation, the mice were randomized on the basis of weight into six groups (n = 5). WT or Dtgif2k-b tachyzoites were prepared in sterile PBS and intraperitoneally injected into the mice, which were then monitored at least 3 times a day. The mice used in this study were housed in American Association for Accreditation of Laboratory Animal Care (AAALAC)-approved facilities at the IUSM Laboratory Animal Research Center (LARC). The Institutional Animal Care and Use Committee (IACUC) at Indiana University School of Medicine approved the use of all animals and procedures (IACUC protocol number 11376).
Quantification and statistical analysis. Quantitative data are presented as the means and standard deviations (SD) from biological triplicates. Statistical significance was determined using one-way analysis of variance (ANOVA) with Tukey's post hoc test or multiple t test two-tailed in Prism (version 7) software (GraphPad Software, Inc., La Jolla, CA). The number of biological replicates (n) and P values are given in the figure legends. For immunoblot analyses, the reported images are representative of at least three independent experiments.
Data availability. RNA-seq data sets from this study are available in the NCBI GEO database (accession number GSE158231).

SUPPLEMENTAL MATERIAL
Supplemental material is available online only.