a gcn2-like eukaryotic initiation factor 2 kinase ... · a gcn2-like eukaryotic initiation factor 2...

EUKARYOTIC CELL, Nov. 2011, p. 1403–1412 Vol. 10, No. 111535-9778/11/$12.00 doi:10.1128/EC.05117-11Copyright © 2011, American Society for Microbiology. All Rights Reserved.

A GCN2-Like Eukaryotic Initiation Factor 2 Kinase Increases theViability of Extracellular Toxoplasma gondii Parasites�†

Christian Konrad,1,2 Ronald C. Wek,2 and William J. Sullivan, Jr.1,3*Departments of Pharmacology and Toxicology,1 Biochemistry and Molecular Biology,2 and Microbiology and

Immunology,3 Indiana University School of Medicine, Indianapolis, Indiana 46202

Received 15 May 2011/Accepted 2 September 2011

Toxoplasmosis is a significant opportunistic infection caused by the protozoan parasite Toxoplasma gondii,an obligate intracellular pathogen that relies on host cell nutrients for parasite proliferation. Toxoplasmaparasites divide until they rupture the host cell, at which point the extracellular parasites must survive untilthey find a new host cell. Recent studies have indicated that phosphorylation of Toxoplasma eukaryotictranslation initiation factor 2-alpha (TgIF2�) plays a key role in promoting parasite viability during times ofextracellular stress. Here we report the cloning and characterization of a TgIF2� kinase designated TgIF2K-Dthat is related to GCN2, a eukaryotic initiation factor 2� (eIF2�) kinase known to respond to nutrientstarvation in other organisms. TgIF2K-D is present in the cytosol of both intra- and extracellular Toxoplasmaparasites and facilitates translational control through TgIF2� phosphorylation in extracellular parasites. Wegenerated a TgIF2K-D knockout parasite and demonstrated that loss of this eIF2� kinase leads to a significantfitness defect that stems from an inability of the parasite to adequately adapt to the environment outside hostcells. This phenotype is consistent with that reported for our nonphosphorylatable TgIF2� mutant (S71Asubstitution), establishing that TgIF2K-D is the primary eIF2� kinase responsible for promoting extracellularviability of Toxoplasma. These studies suggest that eIF2� phosphorylation and translational control are animportant mechanism by which vulnerable extracellular parasites protect themselves while searching for a newhost cell. Additionally, TgIF2� is phosphorylated when intracellular parasites are deprived of nutrients, butthis can occur independently of TgIF2K-D, indicating that this activity can be mediated by a different TgIF2K.

The ability to rapidly respond to stress is essential for cellu-lar survival. Translational control through the phosphorylationof the alpha subunit of eukaryotic initiation factor 2 (eIF2�) isa well conserved mechanism used by cells to repress globalprotein synthesis during times of nutrient scarcity (14, 48). Ina GTP-driven process, eIF2 delivers the initiator tRNA to thetranslational machinery. When GCN2 phosphorylates eIF2� atits regulatory serine residue, this translation initiation factorbecomes an inhibitor of its guanine nucleotide exchange factor,eIF2B, which results in reduced levels of the eIF2-GTP-Met-tRNAi ternary complex that are accessible to the translationalmachinery. Consequently, general protein synthesis is dimin-ished, which conserves cellular resources and provides time forthe cell to reprogram its genome to adapt to the stress. Inaddition to the GCN2 protein kinase that is activated duringnutritional starvation, other eIF2� kinases respond to differentstress conditions (41, 48). These protein kinases includePERK/PEK, activated by endoplasmic reticulum (ER) stress,and HRI and PKR, which respond to heme depletion and viralinfection, respectively.

We have previously shown that the obligate intracellularparasite Toxoplasma gondii (phylum Apicomplexa) relies onphosphorylation of eIF2� (designated TgIF2�) and transla-

tional control to remain viable during the times it must persistwithout host cells (18). Toxoplasma can cause congenital birthdefects, ocular disease, and life-threatening opportunistic in-fection (46). Current treatments consist of antifolates, whichare problematic due to toxicity issues; therefore, there is anurgent need to develop novel therapies to treat this parasiticinfection (6). Phosphorylation of eIF2� has recently beenshown to be critical during multiple phases of the life cycle ofapicomplexan parasites (8, 17, 18, 27, 51). We generated aToxoplasma mutant that no longer phosphorylates TgIF2� bymutating the regulatory serine (Ser71) to alanine (18). TheTgIF2�-S71A mutant suffered a significant fitness defect invitro and in vivo because the mutants were more susceptible toextracellular exposure. Toxoplasma expresses four putativeeIF2� kinases designated TgIF2K-A to -D, and the underlyingprotein kinase mediating this translational control in responseto extracellular stress has not yet been identified (18, 27).

In this study, we hypothesized that extracellular parasitesendure nutrient deprivation through the activity of a GCN2orthologue. We show that TgIF2K-D is a GCN2-like eIF2�kinase in Toxoplasma. Through the generation of TgIF2K-Dmutants, TgIF2K-D was shown to enhance the viability ofparasites during times when they are deprived of their hostcells. Parasites with a knockout of TgIF2K-D are unable tophosphorylate TgIF2� and initiate translational control in re-sponse to extracellular stress, phenocopying the TgIF2�-S71Amutant (18). Intracellular parasites also phosphorylate TgIF2�during nutrient deprivation, but this can occur independentlyof TgIF2K-D, suggesting the involvement of another TgIF2K.This study indicates that TgIF2K-D is the eIF2� kinase facil-

* Corresponding author. Mailing address: Department of Pharma-cology and Toxicology, Indiana University School of Medicine, 635Barnhill Drive, MS A-525, Indianapolis, IN 46202. Phone: (317) 274-1573. Fax: (317) 274-7714. E-mail: [email protected].

† Supplemental material for this article may be found at http://ec.asm.org/.

� Published ahead of print on 9 September 2011.

1403

on May 21, 2020 by guest

http://ec.asm.org/

Dow

nloaded from

http://ec.asm.org/

itating translational control in extracellular parasites, enhanc-ing the survival of Toxoplasma as it searches for a new host cell.

MATERIALS AND METHODS

Parasite culture. Toxoplasma tachyzoites were maintained in human foreskinfibroblasts (HFFs) in Dulbecco modified Eagle’s medium (DMEM) containing25 mM glucose and 4 mM glutamine (Invitrogen) and supplemented with 1%heat-inactivated fetal bovine serum (Gibco/Invitrogen) at 37°C and 5% CO2.

Cloning of the TgIF2K-D cDNA. Tachyzoite mRNA was used to generate acDNA library (Omniscript; Qiagen) for the amplification of the TgIF2K-D openreading frame (ORF). This PCR amplification employed primers specific to theTgIF2K-D gene that was annotated in the Toxoplasma database (www.toxodb.org, TgME49_119610). Tachyzoite mRNA was reverse transcribed using theSuperScript one-step reverse transcription-PCR (RT-PCR) kit (Invitrogen) withrandom and oligo(dT) primers according to the manufacturer’s recommenda-tions. All PCRs were carried out with Phusion DNA polymerase (Finnzymes)using the provided GC buffer. The GeneRacer kit (Invitrogen) was used for the5� and 3� rapid amplification of cDNA ends (RACE) of the TgIF2K-D gene.

Generation of TgIF2K-D knockout parasites. To generate TgIF2K-D knockoutparasites (�if2k-d), we amplified �1.5-kb DNA fragments upstream and down-stream of the start and stop codons carried in the TgIF2K-D locus. Oligonucleo-tide primers used to amplify the 5� flanking sequence were designated 1 and 2,and the primers used to amplify the 3� flanking sequence were designated 3 and4 (see Table S1 in the supplemental material). The amplified DNA was insertedinto the pDHFR*-TSc3 vector (35), such that these fragments flanked opposingends of a modified dihydrofolate reductase-thymidylate synthase (DHFR*-TS)minigene, which confers resistance to pyrimethamine. The resulting knock-out vector was designated �TgIF2K-D::DHFR*. Fifty micrograms of the�TgIF2K-D::DHFR* knockout vector was linearized with NotI and transfectedinto RH strain parasites lacking Ku80 (10, 15) as described previously (35).Transfected parasites were grown in HFF cells in the above-defined DMEMsupplemented with 1 �M pyrimethamine and cloned by limiting dilutions. Indi-vidual parasite clones were screened by PCR to confirm the replacement of theTgIF2K-D genomic locus with the DHFR*-TS minigene.

To confirm that the correct insertion occurred at the TgIF2K-D locus, primerscomplementary to the 3� untranslated region (3�-UTR) of the DHFR* minigene(primer 10) and upstream of the insertion site (primer 9) were used in a PCRassay with genomic DNAs purified from the candidate knockout parasites. PCRassays using primers 5 and 6, which are complementary to exon III, and primers7 and 8, which were used to generate the genetic tagging vector (see below), werecarried out to verify the absence of the complete TgIF2K-D genomic locus. Lossof TgIF2K-D mRNA expression was verified by RT-PCR using primers comple-mentary to sequences upstream (primer 11) and downstream (primer 12) of theencoded protein kinase domain. As control, a portion of Toxoplasma actin(TgME49_009030) mRNA was amplified by RT-PCR using primers 13 and 14.

Genetic tagging of TgIF2K-D. For the expression of TgIF2K-D tagged withhemagglutinin (HA) at its C terminus, a 1.2-kb DNA fragment containing exonXVIII was amplified using Toxoplasma genomic DNA as the template andprimers 7 and 8. The amplified DNA segment was then inserted into the vector3�HA-LIC-DHFR-TS using the ligation-independent cloning method (43).LIC-HA3�-DHFR-TS is a derivative of pYFP-LIC-DHFR (15) in which theyellow fluorescent protein (YFP)-coding fragment had been replaced with threecontiguous HA tags. Fifty micrograms of the TgIF2K-D-HA3� plasmid waslinearized with the restriction enzyme AscI and then transfected into RH�ku80parasites. Following limiting dilutions, positive clones were identified using amonoclonal antibody that specifically recognizes the HA tag (Roche).

The 1.2-kb DNA fragment containing exon XVIII (amplified with primers 7and 8) was also ligated into a LIC-HA2�-DD-DHFR-TS vector (15) to generatea TgIF2K-D fusion with 2�HA and a Shield-regulated destabilization domain(DD) at the C terminus (2�DD). Following transfection of this linearized plas-mid, individual parasite clones were screened for the stabilization of TgIF2K-D2�DD in the presence of Shield-1 (500 nM; Clontech) using the anti-HAmonoclonal antibody.

Parasite motility and invasion assays. Parasites of the designated strain werereleased from host cells via syringe passage and filter purified as described above.Motility of extracellular tachyzoites was assessed by monitoring deposits ofSAG1 antigen as described previously (4). The number of parasites with antigentrails was determined in 10 random microscopic fields in 3 independent exper-iments. Adhesion and invasion assays were performed using differential red/green staining for SAG1 as described previously (16). Attached and invadedtachyzoites were determined in 15 random microscopic fields, and data from 6independent experiments were compiled.

Comparative fitness assay. The comparative fitness assay was carried out asdescribed previously by Joyce et al. (18), with the exception that SYBR green-based quantitative real-time PCR (qPCR) was performed using primers thatspecifically delineated between parental �Ku80 parasites, referred to as wild type(WT), and �if2k-d parasites. In brief, equal numbers of filter-purified parentaland �if2k-d parasites (5 � 105) were cocultured in the same flask of HFF hostcells. At 72 h postinfection, 105 parasites of the mixed population were isolatedfrom the lysed culture and then transferred to a fresh HFF monolayer for anadditional 72 h. This resulted in a total of 6 days of HFF infection by using twoserial passages. Genomic DNA (gDNA) from the parasite samples was isolatedusing the DNeasy kit (Qiagen) and used in qPCR assays. Primers used todistinguish WT from �if2k-d parasites included primers 15 and 16 and primers 17and 18, as indicated. qPCR measurements were normalized by amplifying the5�-UTR of TgIF2K-D, which is present in both WT and �if2k-d parasites (prim-ers 19 and 20). Twenty-five nanograms of gDNA was used in the qPCR assays,which were performed in triplicate using the 7500 real-time PCR system (Ap-plied Biosystems). Relative quantification software (SDS software, version 1.2.1)was used for the analysis. As a specificity control, SYBR green assays employinggDNA purified from either WT or �if2k-d parasites were carried out to verify thespecificity of primers in the qPCR assay (data not shown).

Parasite proliferation assays. Toxoplasma recovery from extracellular stresswas analyzed using standard doubling and plaque assays (35). Parental �ku80(WT), �if2k-d, and TgIF2�-S71A (18) parasites were physically released fromhost cells by syringe passage and then filter purified to remove host cell debris.A total of 106 parasites were subjected to an extracellular stress assay for 0, 8, or10 h in culture medium at 37°C and 5% CO2 without host cells prior to infectingHFF host cells, as described previously (18). Parasites were quantitated using astandard counting assay, with counts performed every 8 h postinfection. Parasitecounting assays were carried out in triplicate using separate biological samples,and results of a representative experiment are shown. In the plaque assays, 500WT, �if2k-d, TgIF2�-S71A, or TgIF2K-D2xDD parasites were used to infect HFFmonolayers in 12-well plates following extracellular incubation for up to 10 h, asindicated. The degree of host cell lysis at 7 days postinfection was determined bycrystal violet staining of methanol-fixed cells. Measurements of the lysed areaswere done using an Alpha Innotech imaging system, and results of a represen-tative experiment of three independent experiments are presented.

Analysis of nutrient starvation of intracellular Toxoplasma. To deprive intra-cellular Toxoplasma of nutrients, we employed a method recently developed byAnthony Sinai (University of Kentucky, unpublished). For these experiments,HFF cells and parasites were maintained in alpha minimal essential medium(�MEM)–7% fetal bovine serum (FBS)–2 mM glutamine (complete medium[CM]) (Gibco). Tachyzoites (106) were allowed to infect HFF monolayers inT-25 flasks for 24 h, at which point medium and extracellular parasites wereremoved by washing with prewarmed Hank’s balanced salt solution (HBSS)lacking glucose. Nutrient starvation was induced by incubating infected HFFs inCM diluted to a 6% final concentration as the starvation medium (6% SM) for8 h; control flasks were incubated in CM. These conditions do not overtly affectthe viability of HFF host cells or induce host cell autophagy (Anthony Sinai,personal communication). Following the 8-h incubation in diluted medium, theinfected monolayer was washed with ice-cold HBSS and parasites were harvestedby syringe passage and filter purified. Analysis of TgIF2� phosphorylation wasperformed by immunoblotting as described previously (44). To monitor therecovery from intracellular nutrient deprivation, 6%-SM was replaced with CMat 4 or 8 h after exposure. The number of parasites per vacuole was counted 24and 36 h later.

Western blotting. Western blot analyses of TgIF2� phosphorylation werecarried out as previously described (18, 27). To analyze the stabilization of theTgIF2K-D2xDD protein, intracellular parasites were grown for up to 24 h inmedium supplemented with 500 nM Shield-1 prior to physical release from thehost cells. HA-tagged proteins were detected by Western blot analyses afterparasite lysates were resolved on a 3 to 8% Tris-acetate polyacrylamide gradientgel. A rat monoclonal antibody that specifically recognizes the HA tag (Roche)was used as a primary antibody, and an anti-rat IgG antibody conjugated withhorseradish peroxidase (GE Healthcare) was used as a secondary antibody.HA-tagged proteins were visualized using a chemiluminescence Western blottingsubstrate (Pierce).

Measurements of protein synthesis. Intracellular parasites were mechanicallyreleased from host cells as described above, and 2.5 � 107/ml tachyzoites weretransferred into Toxoplasma culture medium lacking methionine and cysteine.Labeling was initiated by adding [35S]Cys/Met label (ICN) to a final concentra-tion of 200 �Ci/ml. After 1 h of incubation in DMEM at 37°C and 5% CO2,samples were immediately put on ice. Parasites were harvested by centrifugationat 4°C, and cell pellets were washed twice with ice-cold phosphate-buffered saline

1404 KONRAD ET AL. EUKARYOT. CELL


http://ec.asm.org/

Dow

nloaded from

http://ec.asm.org/

(PBS) and then lysed in 100 �l RIPA buffer (44). Uptake of the 35S during the1-h pulse radiolabeling was similar for the WT and mutant parasites. For eachsample, equal amounts of proteins were precipitated by adding trichloroaceticacid (TCA) to a final concentration of 10%. After incubation on ice for 30 min,samples were collected by centrifugation at 10,000 � g for 30 min at 4°C. TheTCA precipitates were washed twice with acetone and resuspended in an equalvolume of PBS. Incorporation of the radiolabeled amino acids was determinedusing a scintillation counter. Results of all radiolabeling experiments are pre-sented as averages for three independent samples, with P values and standarderrors determined using analysis of variance (ANOVA).

Immunofluorescence assays. HFF monolayers were grown on coverslips, in-fected for 24 h, and then fixed in 3% paraformaldehyde. Immunofluorescenceanalyses using a rat monoclonal antibody that recognizes the HA tag (Roche)followed by goat anti-rat Alexa Fluor 488 as the secondary antibody (Invitrogen)was performed as previously described (29).

RESULTS

Characterization of the GCN2-like kinase TgIF2K-D. Thepredicted gene TgME49_119610 (ToxoDB.org) was previouslydesignated TgIF2K-D and is suggested to encode an ortho-logue of GCN2 (27), the eIF2� kinase that is well-documentedas a responder to nutrient starvation stress in other species (14,48). We used RT-PCR to identify and characterize the full-length TgIF2K-D cDNA. Our analysis revealed a predictedTgIF2K-D product consisting of 2,729 amino acid residues(GenBank accession number JF827031), which modifies thepredicted sequence for TgME49_119610 due to a discrepancyat the exon 3/intron 3 boundary. The predicted start codon forthe TgIF2K-D ORF matches the consensus sequence for trans-lation initiation in Toxoplasma (39) and is preceded by anin-frame stop codon. RACE analyses indicated a 5� untrans-lated region (5�-UTR) of 2,151 bp, which is consistent with thetranscriptional start site (TSS) derived from the Full-parasitesdatabase (49) and chromatin immunoprecipitation-on chip(ChIP-Chip) data available in the ToxoDB, and a 3�-UTR of�1,000 bp (see Fig. S1 in the supplemental material). The5�-UTR was further validated by RT-PCR using primers flank-ing the TSS (see Fig. S2 in the supplemental material).

An alignment between TgIF2K-D and the eIF2� kinasesfrom multiple species was compiled using BLAST andCLUSTALW (see Fig. S3 in the supplemental material).TgIF2K-D (residues 1,318 to 1,630) has the central featurescharacteristic of eIF2� kinases, including an insert betweensubdomains IV and V (Fig. 1; see Fig. S3 in the supplementalmaterial). As judged by BLAST analyses, this portion ofTgIF2K-D is most closely related to putative eIF2� kinasesfrom the parasites Plasmodium falciparum (AAN37036;4e�14) and Trypanosoma brucei (XP_828792.1; 6e�10) (25),followed by characterized GCN2 orthologues from Arabidopsisthaliana (CAD30860; 6e�32) (52), Drosophila melanogaster(AAC13490; 8e�27) (28), Schizosaccharomyces pombe(AAU11313; 2e�25) (50), and Saccharomyces cerevisiae(AAA34636; 1e�22) (47). Another hallmark feature of GCN2is an RWD domain, which is present between residues �800and �1000 of TgIF2K-D, with a significance of 4e�6 as de-termined by the motif search program Pfam (9, 26) (see Fig. S4in the supplemental material). The RWD in GCN2 from S.cerevisiae was reported to directly bind to the activator proteinGCN1 (14, 26), and residue changes in GCN2 that blocked thisbinding, or abolition of the GCN2/GCN1 association by GCN1binding with another RWD-containing protein, YIH1, blocked

GCN2 phosphorylation of eIF2� in yeast depleted of aminoacids (13, 14, 36, 37). Toxoplasma also has a predicted GCN1orthologue (TGME49_031480) and a YIH1-related protein(TGME49_112350), supporting the idea that this networkfunctions to regulate a GCN2-related eIF2� kinase in thisparasite.

The sequences of the histidyl-tRNA synthetase (HisRS) do-main, which stimulates eIF2� kinase activity by binding touncharged tRNAs accumulating during nutrient deprivation(14), appears to be less well conserved in the protozoan GCN2-like kinases. Analysis of the sequences flanking the C-terminalend of the protein kinase domain (residues 1750 to 2360)identified the PRGGRVY2299 sequence as the closest match tothe histidine B sequence (AAGGRYD), which is characteristicfor the HisRS-related domains (42). This weaker conservationof the HisRS-related sequences is a feature shared with otherGCN2-related protein kinases from apicomplexans, includingP. falciparum (8). TgIF2K-D also lacks the pseudokinase do-main found in mammalian and yeast GCN2s, which is thoughtto contribute to the eIF2� kinase activity (33). The C terminusof GCN2 is important for dimerization and ribosome associa-tion (14, 48), and this region in TgIF2K-D (residues 2436 to2499) is rich in hydrophobic and basic residues, which aresuggested to contribute to these regulatory processes in thiseIF2� kinase. Interestingly, this region shares sequence iden-tity with GCN2-like kinases encoded in the apicomplexansNeospora caninum (NCLIV_010550, 3e�30), Cryptosporidiummuris (CMU_027700; 0.011), Plasmodium falciparum (PF14_0264; 9e�08), Plasmodium berghei (PBANKA_101620,4.8e�08), Plasmodium knowlesi (PKH_113740, 1.1e�07), andPlasmodium vivax (PVX_085120; 2e�07) (see Fig. S5 in thesupplemental material). We designated this conserved regionthe C-terminal homology (CTH) region (Fig. 1).

Based on the presence of sequences related to the eIF2kinases juxtaposed to the signature RWD domain, a putativehistidine B-like sequence, and a C terminus rich in hydropho-bic and basic residues, TgIF2K-D is suggested to be a parasiteorthologue of GCN2 (Fig. 1). We therefore hypothesize that

FIG. 1. Domain structure of TgIF2K-D. TgIF2K-D contains a pro-tein kinase domain (black boxes) with an insert (I) characteristic ofeIF2� kinases and signature regulatory regions, including the RWDdomain (dark gray) and a proposed HisRS-related region (light gray).The conserved C-terminal homology (C-term) domain is denoted witha stippled box. The numbers below the diagram demarcate the aminoacid residues for each of the domains of TgIF2K-D. A diagram ofmouse GCN2 (MmGCN2) is shown for comparison.

VOL. 10, 2011 TgIF2K-D PROMOTES RESISTANCE TO EXTRACELLULAR STRESS 1405


http://ec.asm.org/

Dow

nloaded from

http://ec.asm.org/

TgIF2K-D plays a critical role during nutrient deprivation ex-perienced by extracellular Toxoplasma.

TgIF2K-D is expressed in intra- and extracellular parasites.Using RH�ku80 parasites engineered to have greater frequen-cies of homologous recombination (10, 15), we endogenouslytagged TgIF2K-D with three HA epitopes (3�HA) at the Cterminus. Western blot analyses of total protein lysate usinganti-HA antibody identified three clustered protein bands witha molecular mass similar to the deduced 289 kDa forTgIF2K-D (Fig. 2A). These proteins were not present in theuntagged parental parasites referred to as wild-type (WT). Wealso observed that upon extracellular incubation for up to 8 h,a condition that induces high levels of TgIF2� phosphorylation(18) (Fig. 2B), the levels of the faster-migrating TgIF2K-D3�HA variants diminished while those of the slower-migratingprotein increased slightly (Fig. 2A). The difference betweenthese variants of TgIF2K-D may be attributable to a posttrans-lational modification(s), such as protein phosphorylation, orto alternative mRNA splicing, which may contribute toTgIF2K-D activation by stress. While we did not detect alter-native mRNA splice products during our analysis of theTgIF2K-D cDNAs, alternative mRNA splicing was reported in

earlier studies of mammalian GCN2, although its biologicalsignificance is not yet understood (42).

To identify the cellular location of TgIF2K-D, we also car-ried out immunofluorescence microscopy. The HA-taggedTgIF2K-D localized to the parasite cytosol in both intra- andextracellular parasites (Fig. 2C). A cytosolic localization isconsistent with reports on GCN2 in other species (14, 48).

TgIF2K-D facilitates TgIF2� phosphorylation and transla-tional control in extracellular parasites. Extracellular stress isa potent inducer of TgIF2� phosphorylation, and loss of trans-lational control in the TgIF2�-S71A mutant reduced parasiteviability (18). To address whether TgIF2K-D is required tomanage extracellular stress, we generated knockdown andknockout parasite clones in the RH�ku80 background. Theknockdown of TgIF2K-D involved an in-frame fusion of twoHA tags and a 12-kDa destabilization domain (DD) at the Cterminus of the endogenous TgIF2K-D in the RH�ku80 strain(see Fig. S6 in the supplemental material). The parasite clone,designated TgIF2K-D2xDD, allowed tunable expression of theTgIF2K-D protein. In the absence of the stabilizing ligandShield-1, DD-tagged proteins are rapidly degraded (1, 2);TgIF2K-D2xDD parasites cultured without Shield-1 had no de-tectable levels of TgIF2K-D protein as assayed by Westernblot analysis (see Fig. S6 in the supplemental material). Theknockout of TgIF2K-D eliminated the entire genomic locusthrough homologous recombination and allelic replacementwith a modified dihydrofolate reductase-thymidylate synthase(DHFR-TS) minigene, which confers resistance to pyrimeth-amine (Fig. 3A) (5). �if2k-d was verified by PCR analyses ofgenomic DNA purified from pyrimethamine-resistant clones(Fig. 3B). In addition, total RNAs from the parental strain anda �if2k-d knockout clone were isolated for RT-PCR analysis ofthe TgIF2K-D transcript. While TgIF2K-D mRNA was ampli-fied from parental parasites, the corresponding transcript wasnot detected in �if2k-d parasites (Fig. 3C). This parasite clonerepresents the first knockout of an eIF2� kinase in Toxo-plasma.

Next we addressed whether TgIF2K-D is required for in-duced TgIF2� phosphorylation when the parasite is outsidethe host cell. As observed previously (18), parental WT para-sites showed TgIF2� phosphorylation after 8 h of incubation inthe extracellular environment (Fig. 4A). In comparison, therewas minimal TgIF2� phosphorylation in the TgIF2K-D2xDD

knockdown or �if2k-d knockout parasites following extracel-lular exposure (Fig. 4A). To test the specificity of TgIF2K-D inresponding to extracellular stress, we subjected WT and�if2k-d parasites to the calcium ionophore A23187, a knowninducer of ER stress and TgIF2� phosphorylation (27). Asshown in Fig. 4B, the �if2k-d parasites were not defective forTgIF2� phosphorylation in response to ER stress. These re-sults support the model that each TgIF2� kinase in Toxo-plasma recognizes distinct stress arrangements and TgIF2K-Dis central for inducing TgIF2� phosphorylation when parasitesare outside the host cell.

Under stress conditions, eIF2� phosphorylation repressesgeneral translation as part of the cellular stress response (41).To compare translational control in WT versus �if2k-d para-sites, we measured the incorporation of radiolabeled Cys/Metin parasites subjected to extracellular stress for 1 and 8 h. Asexpected, protein synthesis was repressed by greater than 90%

FIG. 2. Size and localization of TgIF2K-D. (A) TgIF2K-D3�HAprotein was detected by Western blotting by probing parasite lysateswith anti-HA antibody. Densitometry values for the slowest-migratingprotein (upper arrow) are listed below the anti-HA blot. The faster-migrating TgIF2K-D3�HA variants (lower arrow) diminished whenparasites were subjected to extracellular stress for 4 or 8 h. Sampleswere normalized in the immunoblot analysis using antibody specific forToxoplasma tubulin. (B) Western blot of TgIF2K-D3�HA parasiteswith antibodies specific for total TgIF2� or phosphorylated TgIF2�(TgIF2��P) during 0, 4, or 8 h of extracellular stress. (C) Immuno-fluorescence analysis using a rat monoclonal HA antibody and ananti-rat Alexa 488 conjugate (green) was performed to show localiza-tion of TgIF2K-D3�HA protein in intra- and extracellular parasites.Nuclear DNA was costained with 4,6-diamidino-2-phenylindole(DAPI) (blue). TgIF2K-D3�HA does not colocalize with nuclear DNA,indicating a cytoplasmic localization in the parasite.



http://ec.asm.org/

Dow

nloaded from

http://ec.asm.org/

in WT parasites subjected to 8 h of extracellular stress; how-ever, in the �if2k-d and TgIF2�-S71A mutant parasites, pro-tein synthesis was diminished by only about 40% (Fig. 4C). Weconclude that TgIF2K-D is likely to be the primary eIF2�kinase that mediates translational control in response to extra-cellular stress.

Parasites lacking TgIF2K-D exhibit a fitness defect. Previ-ously we reported that TgIF2�-S71A mutants are outcompetedby wild-type parasites when placed in a “head-to-head” com-petition assay, as the mutant struggles to cope with the extra-cellular environment experienced while finding a new host cell(18). Given that the �if2k-d mutant failed to phosphorylate

FIG. 3. Generation of a TgIF2K-D knockout. (A) The TgIF2K-D genomic locus, depicted with 18 exons, was replaced by a minigene conferringresistance to pyrimethamine (DHFR*) using homologous recombination in �ku80 RH strain parasites. The numbered arrows indicate thepositions of primers used to screen genomic DNA from transfected pyrimethamine-resistant clones and parental (WT) parasites. Primer sequencesare listed in Table S1 in the supplemental material. (B) Genomic PCR assays used gDNA harvested from WT or �if2k-d parasites and the indicatedprimers to validate replacement of the TgIF2K-D genomic locus. (C) The absence of TgIF2K-D mRNA in the �if2k-d parasites was confirmed byRT-PCR analysis using primers upstream and downstream of the encoded protein kinase domain (primers 11 and 12). Toxoplasma actin mRNAwas amplified as a positive control (primers 13 and 14). A no-template control (Ø) was included in all PCRs.

FIG. 4. TgIF2K-D phosphorylates TgIF2� and represses protein synthesis in response to extracellular stress. (A) Wild-type (WT), TgIF2K-D2xDD (DD), and �if2k-d parasites were exposed for 0 or 8 h to the extracellular environment. TgIF2� phosphorylation was analyzed by separatingcell lysates via denaturing SDS-PAGE, followed by Western blotting using antibodies to total TgIF2� or phosphorylated TgIF2� (TgIF2��P).(B) WT and �if2k-d tachyzoites were treated with 5 �M calcium ionophore A23187 for 30 min and then analyzed for TgIF2��P by immuno-blotting. (C) WT, TgIF2�-S71A, and �if2k-d parasites were physically released from host cells and incubated for 1 or 8 h in DMEM culturemedium. One hour prior to harvesting, the parasites were incubated in the presence of [35S]Cys/Met. Lysates were prepared, and equal amountsof protein were precipitated with TCA. Levels of incorporation of radiolabeled amino acids were determined via scintillation counting. Threeexperiments were performed, and incorporation of the radiolabel is represented as a percentage of that measured for parasites subject to 1 h ofstress. Error bars indicate the standard error, and significance was determined using a two-tailed Student’s t test, with P � 0.05, as indicated bythe asterisks.



http://ec.asm.org/

Dow

nloaded from

http://ec.asm.org/

TgIF2� in response to extracellular stress (Fig. 4A), we testedwhether the �if2k-d parasites would be outcompeted by pa-rental wild-type parasites using the head-to-head fitness assay.Equal numbers of WT and �if2k-d parasites were premixedand transferred into the same culture flask containing a con-fluent monolayer of HFF cells (Fig. 5A). Samples were takenprior to infection and after day 6 for genomic DNA isolation.The relative amounts of WT and �if2k-d parasites were deter-mined using a SYBR green-based quantitative PCR assay andprimers specific for WT or �if2k-d parasites. Primers that am-plify DNAs from both strains were used to ensure normaliza-tion between the samples (Fig. 5B). WT parasites outgrew themutant parasites by day 6 (Fig. 5C), establishing that parasiteslacking TgIF2K-D exhibit reduced fitness in the parasite lyticcycle.

TgIF2K-D promotes the viability of extracellular tachyzoites.We evaluated whether the reduced fitness seen in the �if2k-dparasites involved impaired motility or host cell invasion usinggliding assays and a red/green adhesion and invasion assay. Asreported for the TgIF2�-S71A mutant, �if2k-d parasites ex-

hibited no deficiencies in motility or the ability to attach andinvade host cells (see Fig. S7 and S8 in the supplementalmaterial).

As mentioned above, parasites deficient for TgIF2K-D suffera loss in viability due to an inability to respond appropriately tothe extracellular stress experienced while outside host cells. Tofurther address the role of TgIF2K-D in the extracellular stressresponse, the WT and different mutant parasites (TgIF2�-S71A, �if2k-d, and TgIF2K-D2xDD without Shield) were incu-bated outside host cells in DMEM for between 0 and 10 h priorto being applied to a fresh host cell monolayer. After 7 days,the infected host cells were fixed and stained to determine thedegree of host cell lysis. With increased periods of extracellularstress, �if2k-d parasites showed sharply reduced infection andlysis of host cells that was similar to that measured for theTgIF2�-S71A mutants (Fig. 6A). This defect was more pro-nounced in the �if2k-d parasites than in the TgIF2K-D2xDD

knockdown parasites, suggesting that there are residual levelsof functional TgIF2K-D despite the absence of Shield.

To further characterize the role of translational control inthe resistance to extracellular stress, we also analyzed the dou-bling rate of the �if2k-d parasites. �if2k-d parasites prolifer-ated at a rate similar to that of the WT when allowed to infecta new host cell monolayer immediately upon release from theirinitial host cells (Fig. 6B, 0-h extracellular stress). However,consistent with the plaque assay, extracellular stress led to asignificant reduction in the proliferation of �if2k-d parasites.WT parasites subjected to extracellular stress for 10 h grew toan average of �17 parasites/vacuole, but �if2k-d parasites grewonly to �10 parasites/vacuole (Fig. 6B). This reduction indoubling time was also observed when the TgIF2�-S71A mu-tants were subjected to extracellular stress prior to infection ofthe HFF cells. Collectively, these studies establish thatTgIF2K-D is critical for promoting survival of extracellulartachyzoites through translational control mediated by thephosphorylation of TgIF2�.

TgIF2� is phosphorylated in starved intracellular parasitesin the absence of TgIF2K-D. In addition to nutrient depriva-tion experienced while outside a host cell, parasites can bedeprived of host nutrients while intracellular as well. Toxo-plasma is auxotrophic for several key metabolites and aminoacids, including tryptophan. Tryptophan starvation has beenshown to be a mechanism of gamma interferon (IFN-)-me-diated parasite growth inhibition (31). We tested whether in-tracellular parasites phosphorylated TgIF2� in response tonutrient starvation and whether TgIF2K-D mediated this ac-tivity. We subjected infected HFFs to starvation medium for8 h, which is composed of complete medium diluted to a 6%final concentration in HBSS (6% SM). TgIF2� phosphoryla-tion was detected in parasites incubated in 6% SM for 8 h,indicating that translational control is initiated in response tonutrient deprivation in intracellular parasites (Fig. 7A). The�if2k-d mutants were still able to phosphorylate TgIF2� inresponse to nutrient deprivation in intracellular parasites (Fig.7A), suggesting that a different TgIF2K is activated in starvedintracellular parasites.

To further characterize the role of TgIF2� phosphorylationin intracellular parasites when nutrients become limiting, weanalyzed the recovery of wild-type, �if2k-d, and TgIF2�-S71Amutants following exposure to 6% SM using a standard dou-

FIG. 5. TgIF2K-D contributes to the fitness of Toxoplasmatachyzoites. (A) Schematic of the “head-to-head” fitness assay.(B) Map of primers (arrows) used to distinguish between wild-type(WT) and �if2k-d parasites. Relative levels of WT and �if2k-d para-sites were determined using a SYBR green assay with primers 26 and27 or primers 28 and 29, as indicated. Samples were normalized for theamplification of a DNA fragment carrying the 5�-UTR, which is con-served between WT and �if2k-d parasites (primers 24 and 25). Errorbars indicate standard errors, and significance was determined using atwo-tailed Student’s t test, with P � 0.01, as indicated by the asterisk.



http://ec.asm.org/

Dow

nloaded from

http://ec.asm.org/

bling assay. Wild-type and �if2k-d parasites exhibited no dif-ference in their ability to recover following incubation for 4 or8 h in 6%-SM (Fig. 7B). In contrast, TgIF2�-S71A parasitesexhibited a greater defect in recovering from the intracellularnutrient starvation (Fig. 7B). These data demonstrate thatphosphorylation of TgIF2� promotes the viability of intracel-lular tachyzoites that experience nutrient deprivation but thatthis response can be mediated by a TgIF2K other thanTgIF2K-D.

DISCUSSION

In this study, we generated and characterized the first eIF2�kinase knockout in the obligate intracellular parasite Toxo-plasma. The TgIF2K-D knockout showed reduced TgIF2�phosphorylation and translational control in response to extra-cellular stress, along with reduced viability when outside the

host cell (Fig. 4 and 6). This phenotype was also observed forthe TgIF2�-S71A mutant, supporting the idea that inducedTgIF2K-D phosphorylation of TgIF2� is central for Toxo-plasma to persist in the extracellular environment (Fig. 6) (18).Intracellular tachyzoites proliferate within a parasitophorousvacuole membrane that operates as a molecular sieve andregulates the acquisition of nutrients from the host cell (38,40). Upon exit from their host cell, the tachyzoites must find anew host cell in order to survive and replicate. The extracel-lular environment is likely to be reduced in essential nutrientsthat are available to the parasite, and/or the tachyzoites maynot be equipped with the uptake mechanisms needed to ac-quire them. Our data suggest that TgIF2� phosphorylationserves to protect the parasite during this period of vulnerabil-ity. Reductions in global translation would allow thetachyzoites to conserve energy and nutrients and may also

FIG. 6. TgIF2K-D protects Toxoplasma against extracellular stress. (A) Five hundred wild-type (WT), TgIF2K-D2xDD (DD), �if2k-d, orTgIF2�-S71A (S71A) parasites physically released from host cells were incubated extracellularly in DMEM culture medium for the designatedtimes before being allowed to infect HFF monolayers in 12-well plates. At day 7 postinfection, infected host cells were fixed and stained with crystalviolet, and the degree of host cell lysis was determined using an Alpha Innotech imager. Three independent experiments were carried out, eachin triplicate, and results of a representative experiment are shown as the percentage of the values for corresponding nonstressed parasite strain.(B) Proliferation of wild-type (WT), �if2k-d, and TgIF2�-S71A (S71A) parasites following extracellular incubation was determined by countingthe number of tachyzoites in 50 random vacuoles at the designated time points postinfection. Three independent experiments were carried out,and the average number of parasites per vacuole for one representative experiment is shown. Error bars, representing standard errors, andsignificance were determined using a two-tailed Student’s t test (P � 0.05).



http://ec.asm.org/

Dow

nloaded from

http://ec.asm.org/

induce preferential translation of key proteins required forextracellular survival (e.g., membrane transporters or a newarray of metabolic enzymes). Such preferential translation oftranscripts, such as ATF4 in mammals and GCN4 in S. cerevi-siae, during eIF2� phosphorylation is central for amelioratingnutrient stress (14, 41, 45, 48). The ability of the parasite toovercome extracellular stress is suggested to be important forpathogenesis, as demonstrated by our prior report that theTgIF2�-S71A mutant has reduced virulence when inoculatedinto mice (18). Toxoplasma strains differing in virulence arealso suggested to differ in their ability to initiate translationalcontrol; hypervirulent strains are able to phosphorylateTgIF2� faster and more robustly than hypovirulent strainsduring extracellular stress (18).

Mechanisms by which Toxoplasma copes with the extracel-lular environment. The mechanisms employed by tachyzoitesto overcome the dramatic changes in their extracellular envi-ronment are poorly understood but have recently emerged asa new area of intensive research. Microarray analyses haverevealed significant changes in the transcriptome between in-tra- and extracellular tachyzoites (12, 21). Generally, intracel-lular parasites favor expression of genes involved in metabo-lism and DNA replication, while Toxoplasma cells in theextracellular environment activate genes focused on invasion,motility, and signal transduction.

Coincident with the reprogramming of the transcriptome,extracellular parasites form a novel plant-like vacuole/vacuolarcompartment (PLV/VAC). The PLV/VAC may protect para-sites from osmotic or ionic stresses encountered outside hostcells or mediate the proteolytic maturation of proproteins tar-geted to micronemes, a cellular compartment important forthe parasite invasion into host cells (11, 24, 30). Several studieshave also shown that extracellular parasites undergo a meta-bolic shift from oxidative phosphorylation to glycolysis in orderto generate the ATP required for gliding motility and invasion(22, 32). Collectively, these studies suggest that tachyzoitesundergo extensive changes in their morphology, metabolism,and transcriptome when transitioning to the extracellular en-vironment.

Translational control through TgIF2� phosphorylation pro-

vides an additional mechanism that can modulate Toxoplasmagene expression that is designed to facilitate extracellular sur-vival. In support of this model, our data showed that parasiteslacking the GCN2-like TgIF2K-D are significantly impaired intheir ability to survive outside host cells. In addition toTgIF2K-D, Toxoplasma is suggested to express three othereIF2� kinases that are each proposed to respond to uniquestress arrangements or environmental cues. TgIF2K-A residesin the parasite endoplasmic reticulum and is suggested tofunction analogously to mammalian PEK/PERK (27, 44).TgIF2K-B is a parasite-specific eIF2� kinase likely to respondto a cytosolic stress (27). Finally, TgIF2K-C is another GCN2-like protein kinase present in the Toxoplasma genome (27).However, this putative eIF2� kinase appears to lack an RWDthat was reported to be essential for GCN2 activity in the yeastmodel system (13, 19, 20). We do not yet understand thefunctional significance of two related GCN2 eIF2� kinases inToxoplasma, although this study demonstrates that deletion ofTgIF2K-D alone is sufficient to disrupt the translational con-trol required for the parasite to cope with the extracellularenvironment. It is tempting to speculate that TgIF2K-C isactivated to phosphorylate TgIF2� during nutrient deprivationexperienced by intracellular parasites. Our data show thatwhile TgIF2� is phosphorylated in starved intracellular para-sites, TgIF2K-D is dispensable for this response. It will beimportant to identify the TgIF2K involved in phosphorylatingTgIF2� under intracellular starvation conditions, sinceTgIF2�-S71A is deficient in recovering from this stress. Col-lectively, our data further highlight the eIF2� kinase stressresponse pathway as a potential therapeutic target.

GCN2-like protein kinases in parasites. The tandem ar-rangement of GCN2-related eIF2� kinases is also found in therelated parasite Plasmodium falciparum. Conservation of mul-tiple GCN2-related protein kinases may indicate that eachphosphorylates eIF2� in response to distinct stress conditions.The P. falciparum PF14_0264 product is most closely related toTgIF2K-D and contains an RWD domain, while PfeIK1 ap-pears to lack an RWD domain and has recently been reportedto respond to amino acid starvation during the intraerythrocytering stage (8). This observation suggests that the RWD/GCN1

FIG. 7. Induction of TgIF2� phosphorylation in starved intracellular Toxoplasma. (A) Intracellular wild-type (WT) or �if2k-d parasites wereincubated in CM or 6% SM for 8 h. TgIF2� phosphorylation was analyzed by Western blotting using antibodies to total TgIF2� or phosphorylatedTgIF2� (TgIF2��P). (B) Proliferation of intracellular wild-type (WT), �if2k-d, and TgIF2�-S71A (S71A) parasites following 4 or 8 h of incubationin 6% SM was determined by counting the number of tachyzoites in 50 random vacuoles at the designated time points. Three independentexperiments were carried out, and the average number of parasites per vacuole for one representative experiment is shown. Error bars representstandard errors, and significance was determined using a two-tailed Student’s t test (P � 0.01).



http://ec.asm.org/

Dow

nloaded from

http://ec.asm.org/

regulatory network may not be essential for invoking transla-tional control during periods of certain nutritional deficiencies.

GCN2-like protein kinases lacking the RWD domain are notrestricted to Apicomplexa. Three GCN2-related kinases(IFKA through -C) have been described in Dictyostelium, butonly IFKC possesses an RWD domain (34). Dictyostelium iscapable of developing a fruiting body, a process that is inducedupon nutrient starvation. Although they are involved in regu-lating Dictyostelium development, neither IFKA nor IFKB ap-pears to represent the initial sensor for this stress, supportingthe idea that different GCN2 isoforms sense distinct stressconditions (3, 7). The role of IFKC in this process has not yetbeen studied.

In the case of mammalian GCN2, different mRNA isoformshave been identified, leading to the expression of one GCN2variant lacking the RWD domain (42). The reason for theabsence of this domain is still enigmatic but is likely to affectthe regulation of eIF2� kinase activity. How the different ver-sions of GCN2 protein kinases interplay and respond to stresswill be an interesting topic for future investigation.

Regulation of translation through TgIF2K-independentmechanisms. Our radiolabeling experiments revealed that�if2k-d and TgIF2�-S71A mutant parasites subjected to extra-cellular stress reduce protein synthesis by about 40% (Fig. 4C).While this is much different from the 90% reduction observedin wild-type parasites, the data suggest that other mechanismsof translation regulation occur in addition to TgIF2� phos-phorylation. The reduction in protein synthesis in parasitesincapable of phosphorylating TgIF2� could be due to dimin-ished amino acids required to sustain translation. Further-more, mammalian target of rapamycin complex 1 (mTORC1)is another key player in translation regulation, which is acti-vated and positively regulates translation when nutrients areavailable (23). A putative orthologue of mTOR is encoded inthe Toxoplasma genome (TgTOR, TGME49_116440), suggest-ing that the mTORC1 pathway of translation control is likelyto be conserved in Toxoplasma, but it has not been character-ized to date. Overall, the similar phenotypes observed for�if2k-d and TgIF2-S71A mutant parasites suggest that trans-lational control via TgIF2� phosphorylation is critical duringthis stress response.

ACKNOWLEDGMENTS

We thank Bradley R. Joyce (Indiana University School of Medicine)and Anthony Sinai (University of Kentucky) for helpful discussionsduring the course of this study. We thank David Sibley (WashingtonUniversity) for providing antitubulin antisera, John Boothroyd (Stan-ford University, Palo Alto, CA) for supplying rabbit anti-SAG1, andVernon Carruthers (University of Michigan Medical School) for pro-viding RH�ku80 parasites and LIC vectors.

Support for this research was provided by the National Institutes ofHealth (R21 grant AI084031 to R.C.W. and W.J.S.), R01 GM049164(R.C.W.), and R01 AI077502 (W.J.S.).

REFERENCES

1. Banaszynski, L. A., L. C. Chen, L. A. Maynard-Smith, A. G. Ooi, and T. J.Wandless. 2006. A rapid, reversible, and tunable method to regulate proteinfunction in living cells using synthetic small molecules. Cell 126:995–1004.

2. Banaszynski, L. A., and T. J. Wandless. 2006. Conditional control of proteinfunction. Chem. Biol. 13:11–21.

3. Bowman, R. L., Y. Xiong, J. H. Kirsten, and C. K. Singleton. 2011. eIF2�kinases Control chalone production in Dictyostelium discoideum. Eukaryot.Cell 10:494–501.

4. Dobrowolski, J. M., and L. D. Sibley. 1996. Toxoplasma invasion of mam-

malian cells is powered by the actin cytoskeleton of the parasite. Cell 84:933–939.

5. Donald, R. G., and D. S. Roos. 1993. Stable molecular transformation ofToxoplasma gondii: a selectable dihydrofolate reductase-thymidylate syn-thase marker based on drug-resistance mutations in malaria. Proc. Natl.Acad. Sci. U. S. A. 90:11703–11707.

6. Dubey, J. P. 2008. The history of Toxoplasma gondii—the first 100 years. J.Eukaryot. Microbiol. 55:467–475.

7. Fang, R., Y. Xiong, and C. K. Singleton. 2003. IfkA, a presumptive eIF2alpha kinase of Dictyostelium, is required for proper timing of aggregationand regulation of mound size. BMC Dev. Biol. 3:3.

8. Fennell, C., et al. 2009. PfeIK1, a eukaryotic initiation factor 2alpha kinaseof the human malaria parasite Plasmodium falciparum, regulates stress-response to amino-acid starvation. Malar. J. 8:99.

9. Finn, R. D., et al. 2010. The Pfam protein families database. Nucleic AcidsRes. 38:D211–D222.

10. Fox, B. A., J. G. Ristuccia, J. P. Gigley, and D. J. Bzik. 2009. Efficient genereplacements in Toxoplasma gondii strains deficient for nonhomologous endjoining. Eukaryot. Cell 8:520–529.

11. Francia, M. E., et al. 2011. A Toxoplasma gondii protein with homology tointracellular type Na()/H() exchangers is important for osmoregulationand invasion. Exp. Cell Res. 317:1382–1396.

12. Gaji, R. Y., M. S. Behnke, M. M. Lehmann, M. W. White, and V. B. Car-ruthers. 2011. Cell cycle-dependent, intercellular transmission of Toxo-plasma gondii is accompanied by marked changes in parasite gene expres-sion. Mol. Microbiol. 79:192–204.

13. Garcia-Barrio, M., J. Dong, S. Ulfano, and A. G. Hinnebusch. 2000. Asso-ciation of GCN1-GCN20 regulatory complex with the N-terminus ofeIF2alpha kinase GCN2 is required for GCN2 activation. EMBO J. 19:1887–1899.

14. Hinnebusch, A. G. 2005. Translational regulation of GCN4 and the generalamino acid control of yeast. Annu. Rev. Microbiol. 59:407–450.

15. Huynh, M. H., and V. B. Carruthers. 2009. Tagging of endogenous genes ina Toxoplasma gondii strain lacking Ku80. Eukaryot. Cell 8:530–539.

16. Huynh, M. H., et al. 2003. Rapid invasion of host cells by Toxoplasmarequires secretion of the MIC2-M2AP adhesive protein complex. EMBO J.22:2082–2090.

17. Joyce, B. R., C. Konrad, R. C. Wek, and W. J. Sullivan, Jr. 2011. Translationcontrol is critical during acute and chronic stages of toxoplasmosis infection.Expert Rev. Anti Infect. Ther. 9:1–3.

18. Joyce, B. R., S. F. Queener, R. C. Wek, and W. J. Sullivan, Jr. 2010.Phosphorylation of eukaryotic initiation factor-2� promotes the extracellularsurvival of obligate intracellular parasite Toxoplasma gondii. Proc. Natl.Acad. Sci. U. S. A. 107:17200–17205.

19. Kubota, H., K. Ota, Y. Sakaki, and T. Ito. 2001. Budding yeast GCN1 bindsthe GI domain to activate the eIF2alpha kinase GCN2. J. Biol. Chem.276:17591–17596.

20. Kubota, H., Y. Sakaki, and T. Ito. 2000. GI domain-mediated association ofthe eukaryotic intiation factor 2alpha kinase GCN2 with its activator GCN1is required for general amino acid control in budding yeast. J. Biol. Chem.275:20243–20246.

21. Lescault, P. J., et al. 2010. Genomic data reveal Toxoplasma gondii differ-entiation mutants are also impaired with respect to switching into a novelextracellular tachyzoite state. PLoS One 5:e14463.

22. Lin, S. S., M. Blume, N. von Ahsen, U. Gross, and W. Bohne. 2011. Extra-cellular Toxoplasma gondii tachyzoites do not require carbon source uptakefor ATP maintenance, gliding motility and invasion in the first hour of theirextracellular life. Int. J. Parasitol. 41:835–841.

23. Martin, D. E., and M. N. Hall. 2005. The expanding TOR signaling network.Curr. Opin. Cell Biol. 17:158–166.

24. Miranda, K., et al. 2010. Characterization of a novel organelle in Toxo-plasma gondii with similar composition and function to the plant vacuole.Mol. Microbiol. 76:1358–1375.

25. Moraes, M. C., et al. 2007. Novel membrane-bound eIF2� kinase in theflagellar pocket of Trypanosoma brucei. Eukaryot. Cell 6:1979–1991.

26. Nameki, N., et al. 2004. Solution structure of the RWD domain of the mouseGCN2 protein. Protein Sci. 13:2089–2100.

27. Narasimhan, J., et al. 2008. Translation regulation by eukaryotic initiationfactor-2 kinases in the development of latent cysts in Toxoplasma gondii.J. Biol. Chem. 283:16591–16601.

28. Olsen, D. S., B. Jordan, D. Chen, R. C. Wek, and D. R. Cavener. 1998.Isolation of the gene encoding the Drosophila melanogaster homolog of theS. cerevisiae GCN2 eIF-2a kinase. Genetics 149:1495–1509.

29. Onyango, D. O., et al. 2011. Base excision repair apurinic/apyrimidinic en-donucleases in apicomplexan parasite Toxoplasma gondii. DNA Repair 10:466–475.

30. Parussini, F., I. Coppens, P. P. Shah, S. L. Diamond, and V. B. Carruthers.2010. Cathepsin L occupies a vacuolar compartment and is a protein mat-urase within the endo/exocytic system of Toxoplasma gondii. Mol. Microbiol.76:1340–1357.

31. Pfefferkorn, E. R., M. Eckel, and S. Rebhun. 1986. Interferon-gamma sup-



http://ec.asm.org/

Dow

nloaded from

http://ec.asm.org/

presses the growth of Toxoplasma gondii in human fibroblasts through star-vation for tryptophan. Mol. Biochem. Parasitol. 20:215–224.

32. Pomel, S., F. C. Luk, and C. J. Beckers. 2008. Host cell egress and invasioninduce marked relocations of glycolytic enzymes in Toxoplasma gondiitachyzoites. PLoS pathogens 4:e1000188.

33. Qiu, H., M. T. Garcia-Barrio, and A. G. Hinnebusch. 1998. Dimerization bytranslation initiation factor 2 kinase GCN2 is mediated by interactions of theC-terminal ribosome binding region and the protein kinase domain. Mol.Cell. Biol. 18:2697–2711.

34. Rai, M., Y. Xiong, and C. K. Singleton. 2006. Disruption of the ifkA and ifkBgenes results in altered cell adhesion, morphological defects and a propensityto form pre-stalk O cells during development of Dictyostelium. Differenti-ation 74:583–595.

35. Roos, D. S., R. G. Donald, N. S. Morrissette, and A. L. Moulton. 1994.Molecular tools for genetic dissection of the protozoan parasite Toxoplasmagondii. Methods Cell Biol. 45:27–63.

36. Sattlegger, E., et al. 2011. Gcn1 and actin binding to Yih1: implications foractivation of the eIF2 kinase GCN2. J. Biol. Chem. 286:10341–10355.

37. Sattlegger, E., et al. 2004. YIH1 is an actin-binding protein that inhibitsprotein kinase GCN2 and impairs general amino acid control when overex-pressed. J. Biol. Chem. 279:29952–29962.

38. Schwab, J. C., C. J. Beckers, and K. A. Joiner. 1994. The parasitophorousvacuole membrane surrounding intracellular Toxoplasma gondii functions asa molecular sieve. Proc. Natl. Acad. Sci. U. S. A. 91:509–513.

39. Seeber, F. 1997. Consensus sequence of translational initiation sites fromToxoplasma gondii genes. Parasitol. Res. 83:309–311.

40. Sinai, A. P. 2008. Biogenesis of and activities at the Toxoplasma gondiiparasitophorous vacuole membrane. Subcell Biochem. 47:155–164.

41. Sonenberg, N., and A. G. Hinnebusch. 2009. Regulation of translation ini-tiation in eukaryotes: mechanisms and biological targets. Cell 136:731–745.

42. Sood, R., A. C. Porter, D. Olsen, D. R. Cavener, and R. C. Wek. 2000. Amammalian homologue of GCN2 protein kinase important for translational

control by phosphorylation of eukaryotic initiation factor-2�. Genetics 154:787–801.

43. Stols, L., et al. 2002. A new vector for high-throughput, ligation-independentcloning encoding a tobacco etch virus protease cleavage site. Protein Expr.Purif. 25:8–15.

44. Sullivan, W. J., Jr., J. Narasimhan, M. M. Bhatti, and R. C. Wek. 2004.Parasite-specific eukaryotic initiation factor-2 (eIF2) kinase required forstress-induced translation control. Biochem. J. 380:523–531.

45. Vattem, K. M., and R. C. Wek. 2004. Reinitiation involving upstream ORFsregulates ATF4 mRNA translation in mammalian cells. Proc. Natl. Acad.Sci. U. S. A. 101:11269–11274.

46. Weiss, L. M., and J. P. Dubey. 2009. Toxoplasmosis: a history of clinicalobservations. Int. J. Parasitol. 39:895–901.

47. Wek, R. C., B. M. Jackson, and A. G. Hinnebusch. 1989. Juxtaposition ofdomains homologous to protein kinases and histidyl-tRNA synthetases inGCN2 protein suggests a mechanism for coupling GCN4 expression toamino acid availability. Proc. Natl. Acad. Sci. U. S. A. 86:4579–4583.

48. Wek, R. C., H. Y. Jiang, and T. G. Anthony. 2006. Coping with stress: eIF2kinases and translational control. Biochem. Soc. Trans. 34:7–11.

49. Yamagishi, J., et al. 2010. High-resolution characterization of Toxoplasmagondii transcriptome with a massive parallel sequencing method. DNA Res.17:233–243.

50. Zhan, K., J. Narasimhan, and R. C. Wek. 2004. Differential activation ofeIF2 kinases in response to cellular stresses in Schizosaccharomyces pombe.Genetics 168:1867–1875.

51. Zhang, M., et al. 2010. The Plasmodium eukaryotic initiation factor-2alphakinase IK2 controls the latency of sporozoites in the mosquito salivaryglands. J. Exp. Med. 207:1465–1474.

52. Zhang, Y., J. R. Dickinson, M. J. Paul, and N. G. Halford. 2003. Molecularcloning of an Arabidopsis homologue of GCN2, a protein kinase involved inco-ordinated response to amino acid starvation. Planta 217:668–675.



http://ec.asm.org/

Dow

nloaded from

http://ec.asm.org/

a gcn2-like eukaryotic initiation factor 2 kinase ... · a gcn2-like eukaryotic initiation factor 2...

Documents