MYST Family Lysine Acetyltransferase Facilitates Ataxia Telangiectasia Mutated (ATM) Kinase-mediated DNA Damage Response in Toxoplasma gondii*

The MYST family of lysine acetyltransferases (KATs) function in a wide variety of cellular operations, including gene regulation and the DNA damage response. Here we report the characterization of the second MYST family KAT in the protozoan parasite Toxoplasma gondii (TgMYST-B). Toxoplasma causes birth defects and is an opportunistic pathogen in the immunocompromised, the latter due to its ability to convert into a latent cyst (bradyzoite). We demonstrate that TgMYST-B can gain access to the parasite nucleus and acetylate histones. Overexpression of recombinant, tagged TgMYST-B reduces growth rate in vitro and confers protection from a DNA-alkylating agent. Expression of mutant TgMYST-B produced no growth defect and failed to protect against DNA damage. We demonstrate that cells overexpressing TgMYST-B have increased levels of ataxia telangiectasia mutated (ATM) kinase and phosphorylated H2AX and that TgMYST-B localizes to the ATM kinase gene. Pharmacological inhibitors of ATM kinase or KATs reverse the slow growth phenotype seen in parasites overexpressing TgMYST-B. These studies are the first to show that a MYST KAT contributes to ATM kinase gene expression, further illuminating the mechanism of how ATM kinase is up-regulated to respond to DNA damage.

The function of many proteins is regulated by a variety of post-translational modifications. Lysine acetylation is rapidly emerging as a major post-translational modification in eukaryotic cells for a multitude of proteins beyond histones, in which this post-translational modification was first described (1). Lysine acetyltransferases (KATs) 3 of the MYST family (named for founding members MOZ, Ybf2/Sas3, Sas2, and TIP60) are broadly conserved among all eukaryotes, distinguished by a hallmark acetyltransferase domain that serves as the enzymatic component of several diverse multiprotein complexes. MYST family members were initially characterized as histone acetyltransferases (HATs) involved in epigenetically mediated transcription control but have also been implicated in a wide variety of critical cellular functions, including gene regulation, cell cycle progression, and DNA replication (2).
MYST KATs also play a significant role in the cellular response to DNA damage and apoptosis. TIP60, in particular, has been shown to activate ataxia telangiectasia mutated (ATM) kinase through acetylation of lysine 3016 (3). HeLa cells expressing a "KAT-dead" dominant negative form of TIP60 lacking acetyltransferase activity display defective DNA repair and fail to undergo apoptosis (4). If ATM kinase is not activated by acetylation, the cell fails to activate cell cycle checkpoints through phosphorylation of DNA damage response proteins (5). It is also possible that MYST KATs participate in the activation of ATM kinase gene expression by virtue of their HAT activity, but this has not been tested.
Insight into the evolution of this vital KAT family is lacking because very little information is available regarding MYST proteins in eukaryotic cells of distal origin. Previously, we have found that the protozoan parasite Toxoplasma gondii (phylum Apicomplexa) possesses two MYST KATs, which we have named TgMYST-A and TgMYST-B, that have high similarity to plant MYST KATs (7). Both TgMYST KATs contain an N-terminal chromodomain (CHD) and C2H2 type zinc finger within the MOZ-SAS KAT domain (PF01853). Histone modifications have recently gained much attention in protozoan parasites because they have been linked to modulating critical facets of pathogenesis and have been validated as novel drug targets (6). For example, in Toxoplasma, histone modifications have been associated with gene-regulatory events relevant to the clinically important process of parasite differentiation, the process by which rapidly growing tachyzoites convert into latent encysted forms known as bradyzoites (7).
We have previously characterized the TgMYST-A KAT (8). In this report, we present evidence that links the KAT activity of TgMYST-B to transcription control, parasite proliferation, and the DNA damage response. We find that Toxoplasma possesses an ATM kinase orthologue that is up-regulated by TgMYST-B; this up-regulation is dependent on the KAT activity of TgMYST-B. Moreover, TgMYST-B is recruited to the ATM kinase gene and protects the parasites from DNA damage. We also present evidence that links the up-regulation of ATM kinase to a slowed growth phenotype observed in parasites overexpressing TgMYST-B. Our results emphasize the importance of TgMYST-B in parasite physiology and heighten its attractiveness as a potential drug target.

EXPERIMENTAL PROCEDURES
Parasite Culture and Reagents-Toxoplasma tachyzoites used in this study were RH strain and maintained in human foreskin fibroblasts using Dulbecco's modified Eagle's medium supplemented with 1.0% fetal bovine serum (Invitrogen). Parasites were grown in a humidified CO 2 (5%) incubator at 37°C. Cultures were routinely monitored for Mycoplasma contamination by MycoAlert TM assay (Cambrex Bio Science). Parasites were harvested immediately following lysis of host cell monolayers and purified by virtue of filtration through a 3.0-m filter (9). In some cultures, the ATM kinase inhibitor KU-55933 (Calbiochem catalog number 118500), anacardic acids (Calbiochem catalog number), methyl methanesulfonate (MMS; Sigma), or vehicle control (DMSO) was added to the media.
Parasite Growth Assays-Doubling times were determined as described previously (9). Briefly, T-25 cm 2 tissue culture flasks containing confluent monolayers of host cells were inoculated with 10 5 freshly lysed tachyzoites and incubated. The number of parasites in 50 randomly chosen vacuoles was counted every 8 h. Toxoplasma growth assays based on detection of the parasite-specific B1 gene (10) were carried out as described previously (11). Briefly, 24-well plates were infected with 1,000 parasites/well; each day, genomic DNA from infected wells was harvested using the DNeasy kit (Qiagen) and used in SYBR Green-based quantitative PCR with the 7500 real-time PCR system (Applied Biosystems). The parasite count for a given sample was calculated by interpolation from a standard curve (11).
Antibodies and Western Blot Analysis-Polyclonal antibody was generated in rabbit to a polypeptide sequence corresponding to the C-terminal 100 amino acid residues of TgMYST-B (amino acids 423-523) at Quality Controlled Biochemicals (Hopkinton, MA). Raw antiserum was affinity-purified on the peptide immobilized on Affi-Gel-15 (Bio-Rad). Specificity of the affinity-purified antibody was characterized by immunostaining Western blots containing 20 g of parasite lysate. Antibody to Toxoplasma tubulin (used at 1:1,000) was provided by David Sibley (Washington University, St. Louis, MO) to serve as a protein loading control. Mouse monoclonal antibody to phosphorylated H2AX was from Millipore (05-636) (12). Mouse monoclonal antibody to ATM kinase (2C1 (1A1); used at 1:2,000) was purchased from Abcam (ab78). Appropriate antimouse or anti-rabbit horseradish peroxidase-conjugated secondary antibodies were employed along with the ECL detection system to visualize results (GE Healthcare). Western blotting for all applications was performed using NuPAGE 4 -12% or 10% SDS-polyacrylamide gels using MOPS or MES running buffer (Invitrogen). For ATM kinase Western blots, proteins were separated on 4 -7% Tris acetate gels (Invitrogen).
Cloning, Expression, and Purification of TgMYST-B-RNA ligase-mediated rapid amplification of cDNA ends (RACE) was performed using GeneRacer (Invitrogen). Amplified products were gel-purified, subcloned into TA-TOPO vectors (Invitrogen), and sequenced. Nucleotide sequencing was performed on both strands at the Indiana University Biochemistry Biotechnology Facility (GenBank TM accession number AAZ79483). To express and purify recombinant epitope-tagged TgMYST-B from Toxoplasma, we used a FLAG tag affinity chromatography approach as described previously (13,14). TgMYST-B was cloned into a Toxoplasma expression vector containing the tubulin (TUB) promoter and hypoxanthine-xanthine-guanine phosphoribosyltransferase (HXGPRT) selection cassette (15,16) to make ptub-fTgMYST-B::HX. The TgMYST-B coding sequences were amplified from Toxoplasma cDNA using a long range PCR kit (Qiagen) and primers containing NdeI and AvrII restriction enzyme sites (in italic type). The 5Ј primer also encoded the FLAG epitope tag (underlined): sense, 5Ј-ATAC-CATCATATGAAAATGGACTACAAGGACGACGACGAC-AAGCCTGGCGACTCCGCGTCTCCAGTCTG; antisense, 5Ј-ATACCATCCTAGGTCAGTCTTCTTCTCCCCGCCTC-GGCGGCG. RH⌬HX parasites were stably transfected by electroporating 25 g of the expression vector into 3.0 ϫ 10 7 parasites, and clones were obtained by limited dilution. fTgMYST-B was purified by sonicating parasite pellets in 0.5 ml of 50 mM Tris-HCl (pH 7.4), 150 mM NaCl, 10% glycerol (v/v), and 1% Triton X-100 supplemented with a protease inhibitor mixture (Sigma P8340). The parasite lysate was cleared by centrifuging at 13,000 rpm for 10 min at 4°C prior to mixing with 40 l of equilibrated anti-FLAG M2-agarose affinity resin (Sigma). The resin and lysate were mixed overnight at 4°C and spun at 8,200 ϫ g for 30 s. The supernatant was removed, and the resin was washed three times in 500 l of 50 mM Tris-HCl (pH 7.4), 150 mM NaCl, and 10% glycerol (v/v)) and once with 500 l of 1ϫ HAT assay buffer (50 mM Tris-HCl (pH 8.0), 5% glycerol, 0.1 mM EDTA, 50 mM KCl, 1 mM dithiothreitol, 1 mM phenylmethylsulfonyl fluoride, 10 mM sodium butyrate) at 4°C. fTgMYST-B bound to the resin was used directly in HAT assays as we have done previously (8). To generate the mutant version, fTgMYST-B(E403G), site-directed mutagenesis was performed using the QuikChange II XL kit (Stratagene) according to the manufacturer's instructions. Western blots were performed with anti-FLAG to ensure equal expression between fTgMYST-B(E403G) and fTgMYST-B in the transgenic clones used (data not shown).
HAT Assay-In vitro HAT assay was performed as previously described (17). Briefly, FLAG-tagged recombinant protein was purified from Toxoplasma lysates using anti-FLAG M2 affinity gel (Sigma) for use in an enzymatic HAT reaction containing recombinant H3 peptide as substrate. HAT reactions were incubated at 30°C for 60 min and stopped by the addition of 10 l of 4ϫ NuPAGE loading dye (Invitrogen) containing 2.0 l of ␤-mercaptoethanol. Samples were separated on 4 -12% BisTris gel with MES buffer and transferred to nitrocellulose membrane for immunoblotting with anti-acetyl lysine (Assay Designs/Stressgen catalog number KAP-TF 120H, used at a 1:1,000 dilution). Densitometry values were determined using Kodak 1D version 3.6.3 software (Scientific Imaging System).
Immunofluorescence and Immunoelectron Microscopy-Tachyzoites were allowed to infect human foreskin fibroblast monolayers grown on glass coverslips. Immunofluorescence assays (IFAs) were carried out as described by Bhatti and Sullivan (15), using a 1:100 dilution of affinity-purified anti-TgMYST-B, a 1:1,000 dilution of anti-FLAG antibody (Sigma F7425), or a 1:4,000 dilution of SAG1 as primary antibody. Secondary antibody consisted of anti-rabbit Alexa Fluor 488 or anti-mouse Alexa Fluor 594 (Invitrogen). Extracellular tachyzoites were examined after coating coverslips with 50 l of 0.1 mg/ml poly-L-lysine. Parasites in 3% paraformaldehyde and PBS were inoculated onto the coated coverslips and allowed to sit at room temperature for 15 min before being processed for IFA as outlined above. For IFAs, 0.3 M 4Ј,6-diamidino-2-phenylindole (DAPI) was applied for 5 min at room temperature in the dark as a co-stain. Preparations of Toxoplasma were fixed in 4% paraformaldehyde (Electron Microscopy Sciences) in 0.25 M HEPES (pH 7.4) for 1 h at room temperature and then in 8% paraformaldehyde in the same buffer overnight at 4°C. For immunolabeling, samples were processed as described (18) using undiluted anti-MYSTB antibodies, subsequently revealed by 10-nm protein A-gold particles before examination with a Philips CM120 electron microscope.
Real-time Reverse Transcription-PCR-Primers were designed using the Primer Express 2.0 software (Applied Biosystems) and are listed in Supplemental Table S1. The reverse transcription reaction was performed using 1.0 g of total RNA isolated from tachyzoites, oligo(dT) primers, and Omniscript reverse transcriptase (Qiagen) according to the manufacturer's directions. 1 l of a 1:10 dilution of the resulting cDNA was amplified in a 25-l total volume containing SYBR Green PCR Master Mix (Applied Biosystems, CA) and 0.5 M of each forward and reverse primer. The Toxoplasma ␤-tubulin gene (GenBank TM accession number M20025) was used to normalize the real-time reverse transcription-PCR (18). All reactions were performed in triplicate, using the 7500 real-time PCR system and analyzed SDS software version 1.2.1 (Applied Biosystems).
Chromatin Immunoprecipitation (ChIP)-ChIP was performed on tachyzoites stably expressing fMYST-B using polyclonal anti-FLAG antibody (Sigma F7425) immobilized to Dynabeads Protein A (Invitrogen). Quantitative PCR was performed as described above. Immunoprecipitated DNA samples were normalized using a standard curve created with serially diluted input DNA. 0.1 ng of total ChIP DNA was added to each reaction, and reactions were performed in triplicate. Primers used for this study are listed in supplemental Table S1.

Characterization of a Second MYST Family KAT in
Toxoplasma-Previously, we reported a partial coding sequence for a second MYST family KAT in Toxoplasma, termed TgMYST-B (GenBank TM accession number AAZ79483) (8). We have used 5Ј-and 3Ј-RACE to acquire the full-length sequence (supplemental Fig. S1). A single amplicon from a nested 5Ј-RACE reaction was sequenced to reveal an in-frame ATG and 5Ј-UTR of 133 nt. Interestingly, the 5Ј-UTR contains a single intron of 358 nt. The 3Ј-RACE product confirmed the previously reported stop codon and revealed the 533-nt 3Ј-UTR. The TgMYST-B genomic locus is 4.7 kb, containing seven introns and eight exons. Our results indicate that the coding sequence of TgMYST-B is 1,572 nt, which is in agreement with the ToxoDB 3.0 prediction, TgTwinScan_2400, not the latest 3,264-nt gene prediction at ToxoDB (TGGT1_ 023320). The discrepancy between the TGGT1_023320 prediction and our cloning results resides at the 5Ј-end of the gene. Consequently, we performed reverse transcription-PCRs using primers designed to clarify the sequence of the 5Ј-end of TgMYST-B. Although products downstream of the transcriptional start site we determined by 5Ј-RACE could be amplified, no amplicons could be generated upstream of this (data not shown). Additionally, no expressed sequence tags or proteomics data to date support the longer TGGT1_023320 prediction.
According to our cloning results, the deduced amino acid sequence of TgMYST-B yields a protein composed of 523 amino acids with a predicted molecular mass of ϳ60 kDa. Protein structure and motif analysis shows that TgMYST-B possesses the expected domains, including the MYST catalytic domain, and an upstream C2H2 zinc finger and CHD (Fig. 1A). BLASTp analysis reveals that the closest matches are MYST acetyltransferases from fellow apicomplexan parasite Cryptosporidium spp. and plant species including Ricinus communis (castor bean) and Arabidopsis thaliana (Ͻe Ϫ86 ). In a phylogenetic analysis, TgMYST-B groups with a MYST KAT in Cryptosporidium parvum, whereas TgMYST-A is more similar to a MYST KAT in Plasmodium species (Fig. 1B). Protein sequence alignments between TgMYST KATs and those found in other select species are shown in supplemental Fig. S2. The aforementioned MYST KAT motifs are well conserved between TgMYST-A and -B, but the intervening sequences and N-terminal sequence were less so (supplemental Fig. S3). Additionally, there are two insertions in TgMYST-B that are not present in TgMYST-A between the CHD and zinc finger/MYST domain; the first is basic-rich and composed of 55 amino acids, and the second is composed of 24 amino acids.
Affinity-purified, polyclonal antibody generated against the C-terminal 100 amino acids (amino acids 423-523) recognizes a single protein of the expected size (ϳ60 kDa) on an immunoblot containing Toxoplasma tachyzoite lysate; no cross-reactivity to either form of TgMYST-A (49 -53 kDa) was detected (Fig. 1C). The protein size supports our cloning data for the true TgMYST-B transcript.
TgMYST-B Displays HAT Activity but Is Found Predominantly in the Cytoplasm-We sought to confirm that TgMYST-B exhibited the expected enzymatic activity using an in vitro HAT assay as described previously (17). Native TgMYST-B is expressed at very low levels, making it difficult to obtain sufficient quantities of enzyme for analysis. Therefore, we engineered a transgenic clone stably expressing an ectopic form of TgMYST-B under control of the strong Toxoplasma tubulin promoter; the recombinant TgMYST-B was tagged with FLAG at the N terminus (fTgMYST-B). fTgMYST-B was purified from parasite lysate using anti-FLAG resin. fTgMYST-B served as the enzyme source in an in vitro HAT assay that uses recombinant H3 as substrate. After incubation, the HAT reaction was resolved on SDS-PAGE for immunoblotting. Acetylated H3 was detected with antibody raised against acetylated lysine. The results verify that TgMYST-B is capable of acetylating histone substrate in vitro ( Fig. 2A). The point-mutated version of TgMYST-B (E403G) is described in detail below.
To determine the subcellular localization of TgMYST-B, we used our affinity-purified polyclonal antibody in IFAs of intracellular Toxoplasma. Results indicate that TgMYST-B appears to be predominantly cytoplasmic with virtually no protein detectable in the nucleus or apicoplast in either intracellular or extracellular tachyzoites (Fig. 2B). Predominantly cytosolic distribution was also seen in transgenic parasites stably expressing ectopic fTgMYST-B (Fig. 2C). For better resolution, we used the anti-MYST-B antibody in immune electron microscopy of intracellular wild type tachyzoites. Density (gold particles/ m 2 ) of labeled structures was determined from 22-26 cellular cryosections. Percentage of cytosolic and intranuclear density was determined from the sum of gold density normalized for the variation in expression of TgMYST-B. Results confirm the cytosolic location for the majority of TgMYST-B (Fig. 2D and supplemental Fig. S4) but also revealed 20% of gold particles to be in the nuclei of the parasites, consistent with the well characterized HAT activity for a member of the MYST family.

TgMYST-B Protects against DNA Damage Induced by
Alkylation-The mammalian MYST HAT TIP60 has been linked with the DNA damage response (19). We found that parasites containing extra TgMYST-B are significantly more resistant to the alkylating DNA-damaging agent MMS (Fig.  3A). To test if the protection against MMS is dependent on the KAT activity of TgMYST-B, we stably expressed a point mutant (E403G) version deficient in catalytic activity (20) (Fig. 2A). Parasites expressing the mutant form of TgMYST-B are no longer protected from MMS (Fig. 3A). At the higher level of MMS (500 M), the parasites harboring mutant TgMYST-B are even more susceptible to DNA damage than wild type, suggesting that the mutant is exerting a dominant negative effect as previously reported for TIP60 (4) and further supporting a link between TgMYST-B and the DNA damage response.
The DNA damage response in eukaryotic cells is strongly correlated with an increased level of histone H2AX phosphorylation (21). H2AX is rapidly phosphorylated in response to double-stranded breaks (termed ␥H2AX), which are induced by MMS. We have established that Toxoplasma contains H2AX that becomes phosphorylated in response to double- stranded DNA breaks (12). In TgMYST-B-overexpressing parasites, however, we found high levels of basal ␥H2AX and found that the increase in ␥H2AX is dependent on the KAT activity of TgMYST-B (Fig. 3B). The higher basal level of ␥H2AX is likely to be a contributing factor explaining why the TgMYST-Boverexpressing parasites are more resistant to MMS.
KAT Activity of TgMYST-B Increases Levels of ATM Kinase-We hypothesized that the increase in ␥H2AX may be due to inappropriately activated ATM kinase. In human cells, ATM kinase responds to genotoxic stresses that lead to doublestranded DNA breaks following activation by TIP60 (22). Toxoplasma possesses a predicted ATM kinase (50.m03243/ TGGT1_024330), and we examined ATM kinase protein levels by immunoblotting Toxoplasma lysate with a cross-reactive antibody. ATM kinase was only detectable in the fTgMYST-Boverexpressing parasites; we could not detect ATM kinase in wild type or parasites expressing mutant fTgMYST-B (Fig. 3C).
To test if the increase in ATK kinase may be due to increased transcript levels, we analyzed ATM kinase mRNA levels in wild type versus parasites overexpressing fTgMYST-B or the mutant version fTgMYST-B(E403G). We detected increased levels of ATM kinase mRNA in fTgMYST-B-expressing parasites compared with wild type and transgenics expressing mutant fTgMYST-B (Fig. 4A). Actin was assayed as a control to show that gene expression is not globally upregulated in the TgMYST-B-overexpressing clone.
We then determined if fTgMYST-B was being recruited to the ATM kinase promoter, which would be consistent with the idea that TgMYST-B activates ATM kinase gene expression. Four regions spanning the predicted start codon of the ATM kinase locus were selected for analysis by ChIP. The results display the expected enrichment of fTgMYST-B in regions most proximal to the predicted start site for ATM kinase (Fig. 4B). Identical ChIP assays were performed on parasites expressing the mutated fTgMYST-B(E403G). fTgMYST-B(E403G) shows a similar localization pattern at the ATM kinase gene, indicating that KAT activity does not interfere with TgMYST-B recruitment to target genes. To ensure that fTgMYST-B is not being recruited to promoters randomly, we performed ChIP at the Toxoplasma actin promoter because actin mRNA is not increased in the fTgMYST-B-overexpressing parasites (Fig. 4A). In contrast to the ATM kinase gene, the presence of fTgMYST-B does not significantly increase at genes like actin that are not up-regulated (Fig. 4B). Collectively, these data strongly suggest that the KAT activity of TgMYST-B protects cells from DNA damage by contributing to the activation of the ATM kinase gene.
TgMYST-B KAT Activity Is Associated with Slowed Replication-It was observed that the parasites expressing additional TgMYST-B were taking considerably longer to lyse their host cell monolayer than wild type parasites. Toxoplasma growth assays comparing wild type and fTgMYST-B-expressing parasites confirmed a significant decrease in parasite proliferation (Fig. 5A). Additional clones isolated from an independent transfection with fTgMYST-B also displayed the same retardation in growth (data not shown). The doubling time of wild type parasites is ϳ8 h, but it increases to ϳ16 -18 h in the fTgMYST-B overexpressor (Fig. 5B). Toxoplasma expressing The designated recombinant FLAG-tagged protein was purified from parasite lysate over anti-FLAG resin and used as the enzyme source in a HAT assay with recombinant H3 peptide substrate as described previously (17). Acetylation was detected using an acetyl-lysine antibody on immunoblots of the HAT assay. Lane 1, fTgGCN5B (used as a positive control); lane 2, fMYST-B; lane 3, wild type lysate; lane 4, fMYST-B(E403G). The -fold induction of H3 acetylation (listed below the gel) was determined by densitometry of the autoradiogram. B, localization of native TgMYST-B by IFA in intracellular (top) and extracellular (bottom) tachyzoites, using anti-TgMYST-B at 1:100 (green). For reference, DNA is stained with DAPI (blue), and for intracellular parasites, the surface antigen SAG1 is stained with anti-SAG at 1:4,000 (red). C, IFA of intracellular tachyzoites engineered to stably express recombinant TgMYST-B tagged with FLAG epitope at the N terminus (shown in green, contrasted with DAPI (blue) and SAG1 (red) for reference). D, immunoelectron microscopy using anti-TgMYST-B performed on intracellular tachyzoites showing gold particles both in the cytoplasm and nucleoplasm (circled particles). n, nucleus; ne, nuclear envelope (dotted line); dg, dense granule; m, microneme.
fTgMYST-B(E403G) grow at the same rate as wild type. The slowed growth is not a result of conversion to the bradyzoite stage as determined by Dolichos lectin staining and PCR for the bradyzoite-specific marker gene BAG1 (data not shown). We conclude that the acetyltransferase activity of TgMYST-B is important in properly regulating growth rate in Toxoplasma tachyzoites.

Inhibition of ATM Kinase or KAT Activity Reverses the Slow Growth Defect in TgMYST-B-overexpressing
Parasites-To explore the possibility that inappropriately activated ATM kinase contributes to the slowed replication of the fTgMYST-B-overexpressing parasites, we added KU-55933, a well characterized ATM kinase inhibitor with an IC 50 of 13.0 nM (23), to the cultures. As shown in Fig. 6A, inclusion of 10 nM KU-55933 restores the growth rate of fTgMYST-B-expressing parasites to wild type levels. Because we have established that excess KAT activity is responsible for the slowed growth of fTgMYST-Boverexpressing parasites (Fig. 5), we examined the effect of pharmacological inhibition of in vivo KAT activities. The addition of anacardic acids (24) also restored the growth rate of fTgMYST-B-overexpressing parasites to levels similar to wild type (Fig. 6B). These data suggest that the mechanism of the reduced replication observed in TgMYST-B-

DISCUSSION
Here we report the full-length cloning and characterization of the second MYST family KAT in the parasite T. gondii, building from a partial clone first described by Smith et al. (8). 5Ј-RACE for TgMYST-B revealed an unusual 5Ј-UTR that harbors an intron, which has been reported for other genes in Apicomplexa (e.g. Plasmodium HGPRT (25)). Similar to TgMYST-A, TgMYST-B can be classified as a "MYST ϩ CHD" KAT, which includes yeast Esa1, human TIP60, and human and Drosophila MOF (2). Both TgMYST-A and -B are similar in size (between 50 and 60 kDa) and subcellular localization and bear the strongest homology to MYST family proteins found in plants.
We present several lines of data suggesting that a subset of total TgMYST-B protein must gain access to the parasite nucleus. Although difficult to detect by IFA, immune electron microscopy shows that 20% of TgMYST-B localized to the nuclear compartment (Fig. 2D). TgMYST-B contains a basicrich stretch of amino acids beginning at residue 125, which is characteristic of a nuclear localization signal (15). Most convincingly, we are able to purify TgMYST-B associated with genomic DNA via ChIP (Fig. 4B). It is intriguing, however, that the bulk of protein is in the cytosol; similar results were observed for TgMYST-A (8). Moreover, we did not detect by IFA any shift in the localization pattern following alkaline pH stress or MMS treatment (data not shown). In other species, MYST KATs are largely nuclear, presumably operating primarily as histone acetyltransferases. However, human TIP60 has been shown to be translocate into the cytoplasm to form "speckles" that facilitate the stabilization of Nmi (N-Myc and STAT interactor) protein (26). In another example, an alternatively spliced form of TIP60 called TIP60␤ is present in both nuclear and cytosolic compartments (27). Toxoplasma is the first cell reported to have such high proportions of MYST KAT protein in the cytosol relative to the nucleus. We speculate two possible reasons for the unusual distribution of TgMYST KATs. First, the distribution may simply be for regulatory reasons to control the amount of TgMYST protein in the nucleus; however, a more tantalizing possibility is that the TgMYST KATs regulate non-histone proteins via acetylation. Several proteomic studies have recently revealed that acetylation is a widespread post-translational modification in prokaryotes and eukaryotes (28,29). Yeast proteome microarray studies have shown that the nucleosome acetyltransferase of H4 (NuA4) complex, which contains MYST KAT Esa1, acetylates cytosolic substrates, such as Pck1 (phosphoenolpyruvate carboxykinase) (30).
Previously, when we attempted to overexpress TgMYST-A in Toxoplasma, we could not do so unless the catalytic domain was mutated (8). Here we report that expression of additional TgMYST-B, but not a mutant form, dramatically slows the growth of transgenic tachyzoites. This result suggests that the KAT activity of TgMYST-B may be important in the regulation of cell cycle, which could be connected to the established role of MYST KATs in the DNA damage response (19). TgMYST-B  . DMSO was used as a vehicle control. The growth rate was monitored using a B1 assay at day 5. **, a significant difference (p ϭ 3.23545 ϫ 10 Ϫ5 ) as determined by t test for unequal variance.
overexpressors show increased protection against the alkylating agent MMS (Fig. 3A). The presence of ATM kinase explains the increased basal levels of ␥H2AX in the TgMYST-B-overexpressing parasites. In turn, increased ␥H2AX causes cell cycle arrest and a decrease in the number of cells in mitosis (31), thus explaining why parasites containing extra TgMYST-B exhibit retarded growth in culture.
We postulated two mechanisms how TgMYST-B may increase the levels of ATM kinase. One mechanism has already been established in humans; TIP60 activates ATM kinase through direct acetylation of Lys 3016 (3,32). Although Toxoplasma ATM kinase contains the conserved lysine residue (Lys 1966 ), we could not test this possibility because the anti-ATM kinase antibodies do not immunoprecipitate the Toxoplasma ATM kinase. In this report, we present evidence for a novel second mechanism that involves TgMYST-B increasing ATM kinase mRNA levels. Further analysis showed that TgMYST-B is specifically recruited to the ATM kinase promoter and that this recruitment is independent of KAT activity. The recruitment of enzymatically active TgMYST-B to the ATM kinase gene correlates with an increase in ATM kinase mRNA levels; conversely, KAT-deficient TgMYST-B, despite being recruited to the ATM kinase gene, is unable to up-regulate mRNA levels. These data strongly suggest that TgMYST-B KAT activity is critical for ATM kinase-mediated DNA repair in Toxoplasma. Our study does not rule out other possibilities that could contribute to increase ATM kinase protein in the TgMYST-B-overexpressing parasites. For example, it is possible that KAT activity of TgMYST-B directly or indirectly leads to stabilization of ATM kinase protein.
Toxoplasma is now the earliest eukaryote demonstrated to have a link between MYST KATs and the response to DNA injury. These findings also suggest that TgMYST-B is orthologous to mammalian TIP60, which is known to be critical for DNA damage response (19). Ongoing studies are attempting to elucidate the TgMYST-B complex to determine the identity of the associating proteins.
Toxoplasma causes congenital birth defects and life-threatening disease in immunocompromised patients. More tolerable drug therapies are urgently needed because the current treatment of pyrimethamine and sulfadiazine has toxic side effects. The data we present here, considered with our previous findings reported by Smith et al. (8), strongly suggest that MYST KATs are involved in diverse critical functions in the parasite, therefore underscoring this protein family as important candidates for future drug design.