Human T-cell Leukemia Virus Type 1 Tax Oncoprotein Prevents DNA Damage-induced Chromatin Egress of Hyperphosphorylated Chk2*

De novo expression of human T-cell leukemia virus type 1 Tax results in cellular genomic instability. We demonstrated previously that Tax associates with the cell cycle check point regulator Chk2 and proposed that the inappropriate activation of Chk2 provides a model for Tax-induced loss of genetic integrity (Haoudi, A., Daniels, R. C., Wong, E., Kupfer, G., and Semmes, O. J. (2003) J. Biol. Chem. 278, 37736–37744). Here we provide an explanation for how Tax induces some Chk2 activities but represses others. We show that Tax interaction with Chk2 generates two activation signals in Chk2, oligomerization and autophosphorylation. However, egress of Chk2 from chromatin, normally observed in response to ionizing radiation, was repressed in Tax-expressing cells. Analysis of chromatin-bound Chk2 from Tax-expressing cells revealed phosphorylation at Thr378, Ser379, Thr383, Thr387, and Thr389. In contrast, chromatin-bound Chk2 in the absence of Tax was phosphorylated at Thr383 and Thr387 in response to ionizing radiation. We further establish that Tax binds to the kinase domain of Chk2. Confocal microscopy revealed a redistribution of Chk2 to colocalize with Tax in Tax speckled structures, which we have shown previously to coincide with interchromatin granules. We propose that Tax binding via the Chk2 kinase domain sequesters phosphorylated Chk2 within chromatin, thus hindering chromatin egress and appropriate response to DNA damage.

Human T-cell leukemia virus type 1 (HTLV-1) 3 is the causative agent of adult T-cell leukemia, a malignancy of CD4 ϩ T lymphocytes (1)(2)(3)(4), and is also known to act in the etiology of a neurodegenerative disease, tropical spastic paraparesis/HTLV-1-associated myelopathy (5,6). The mechanism of leukemogenesis or the neoplastic growth of HTLV-1-infected CD4 ϩ T-cells is incompletely understood. However, a preponderance of experimental data have established that the viral protein Tax is a crucial component in HTLV-1-mediated transformation.
An overall increase in cellular genomic instability is thought to be a driving force behind carcinogenesis (7)(8)(9). Thus, it is not unexpected that HTLV-1-transformed lymphocytes demonstrate a range of chromosomal abnormalities. However, the observation that HTLV-1-infected but not -transformed T-cells also display significant genomic instability (10 -14) suggests that loss of genomic integrity is an event that predisposes the cell to change. More specifically, the observation that Tax expression alone results in genomic instability provides a compelling cause-effect model for the development of adult T-cell leukemia (12,15,16).
Although Tax regulates a wide range of cellular functions, there is no evidence that Tax directly interacts with DNA in a manner that would result in DNA damage (11). Thus, most theories propose that Tax impairs the cellular DNA damage repair response, resulting in increased surviving mutation frequency (15,17,18). Possible molecular models include the repression of human telomerase (10) and ␤-polymerase (19), overactivation of proliferating cell nuclear antigen (20,21), and persistence of unprotected DNA breaks (22). Additionally, Tax has been proposed to perturb the mitotic spindle checkpoint and to directly effect chromosomal segregation, resulting in aneuploidy (23). Another possible mechanism proposes that disruption in the cytokinesis nuclear division coordination is via premature activation of the anaphase-promoting complex achieved by binding of Tax to the anaphase-promoting complex-associated protein Cdc20 (24). In fact, a reasonable hypothesis is that cell cycle ablation, repair enzyme repression, and disruption in the regulation of cell division all contribute to genomic instability. Recently, in an elegant ex vivo approach, Sibon et al. (25) demonstrated that, early in infection, CD4 ϩ T-cells display striking genomic instability and cell cycle redistribution. These authors reported an increase in the number of HTLV-1-infected cells residing in G 2 /M phase, a result consistent with early observations in Tax-expressing cells (26 -28).
We demonstrated previously that Tax expression results in the accumulation of cells in 4N via inappropriate activation of the Chk2-regulated G 2 /M checkpoint (26). Because Chk2 plays a critical bridging role between DNA repair, cell cycle, and apoptosis, a disruption in the intact spatiotemporal relationship of Chk2 activities could result in genomic instability. Consequently, Chk2 is considered to be a tumor suppressor, and Chk2 mutations are associated with an increasing range of clinical scenarios such as neoplasia associated with Li-Fraumeni syndrome, myelodysplastic syndromes, acute myeloid leukemia, lung cancer, osteosarcoma, and breast cancer (29 -34). Thus, the physical association of Tax with Chk2 and the subsequent damage-independent phosphorylation of Chk2 present a valid model for Tax-induced genomic instability. In this study, we demonstrate that Tax expression stabilizes phosphorylated Chk2 protein. This population of phosphorylated Chk2 is sequestered into Tax nuclear complexes and is impaired in its ability to egress from cellular chromatin following ionizing radiation. Thus, it appears that Tax expression results in Chk2 that is able to generate activation of some of its cognate downstream targets, as is reflected in cellular G 2 /M accumulation, but that is impaired in its overall response to DNA damage.

EXPERIMENTAL PROCEDURES
Cell Culture and Transfection-293T cells were maintained at 37°C in a humidified atmosphere of 5% CO 2 in air in Iscove's modified Dulbecco's medium supplemented with 10% fetal bovine serum and 1% penicillin/streptomycin (Invitrogen). Transfections were performed by standard calcium phosphate precipitation. Cells were plated in 150-mm plates at 4 ϫ 10 6 cells/plate. The following day, 20 g of plasmid DNA in 2 M CaCl 2 and 2ϫ HEPES-buffered saline was added dropwise to cells in fresh medium. Cells were incubated at 37°C for 5 h, and fresh medium was added. The cells were harvested 48 h later following a single wash with 1ϫ phosphate-buffered saline (PBS) in 500 l of M-PER mammalian protein extraction reagent (Pierce) with protease inhibitor mixture (Roche Applied Science) and immediately frozen at Ϫ80°C.
Plasmid Construction-The S-tagged expression vectors provide a fusion of a 15-residue peptide (KETAAAKFERQH-MDS) derived from pancreatic ribonuclease A. S-Tax-green fluorescent protein (GFP) was constructed by inserting the taxenhanced GFP fusion open reading frame into the SmaI site of pTriEx4-Neo (Novagen) in-frame with the N-terminal S and His tags. Full-length N-terminally Xpress-tagged Chk2 and its deletion mutants were generated by cloning the PCR-amplified sequences into pcDNA4/HisC (Invitrogen). Full-length Chk2 and deletion mutant sequences (forkhead-associated (FHA) domain, ⌬FHA, and 1-221) were isolated by EcoRI restriction digestion from plasmids pGEX4T-Chk2, pGEX4T-FHA, pGEX4T-⌬FHA, and pGEX4T-1-221 (kindly provided by Dr. David F. Stern, Yale University School of Medicine, New Haven, CT), respectively, and cloned into EcoRI-digested pcDNA4/HisC. Sequences encoding the mutant lacking the serine/threonine catalytic domain and the mutant containing the kinase domain alone were PCR-amplified from the pcDNA4-Chk2 plasmid and cloned in the EcoRI and XhoI restriction sites of pcDNA4/HisC. S-tagged Chk2 was generated by PCR-amplifying full-length Chk2 from pGEX4T-Chk2 and cloning into the EcoRI restriction site of pTriEx4 (Novagen). Hemagglutinin (HA)-Chk2 was also received from Dr. Stern. The generated recombinant DNA constructs were verified by sequencing.
Co-immunoprecipitation-Protein extracts from 2.5 ϫ 10 6 cells transfected with the designated expression vectors were lysed in 500 l of M-PER protein lysis buffer (Pierce) containing the Complete mini-mixture of protease inhibitors (Roche Applied Science) for 30 min at 4°C. Immunoprecipitation was carried out by incubating cell lysates with the appropriate antibody overnight at 4°C on a rotator. 100 l of a 30% slurry of protein A-Sepharose beads (Upstate, Lake Placid, NY) in lysis buffer was added to this mixture, followed by incubation for 3 h at 4°C. The immune complexes bound to beads were pelleted by centrifugation at 3500 rpm for 5 min at 4°C. The beads were washed three times each with 5% sucrose, 1% Nonidet P-40, 500 mM NaCl, 50 mM Tris (pH 7.4), and 5 mM EDTA and with 5 mM Tris-HCl (pH 8.0), 150 mM NaCl, 1% Nonidet P-40, 0.5% sodium deoxycholate, and 0.1% SDS. The bound proteins were eluted by incubation in Laemmli sample buffer at 95°C for 5 min, placed on ice for 1 min, and further centrifuged at 3500 rpm for 2 min. A portion (30 l) of the supernatant was loaded, separated by 10% SDS-PAGE, and transferred to Immobilon-P membranes (Upstate). All subsequent steps were the same as described for Western blotting.
Immunofluorescence Confocal Microscopy-HEK293 were seeded at 1 ϫ 10 5 cells/well on ethanol-washed coverslips in 6-well plates. If appropriate, each well was transiently transfected with the indicated expression plasmids. Cells were washed with PBS and subsequently fixed in 4% paraformaldehyde, permeabilized with methanol, and incubated overnight at 4°C with primary antibody in 3% bovine serum albumin and PBS at a dilution of 1:500. Cells were washed twice with PBS/ Tween 20 and twice with PBS and then incubated for 1 h at room temperature with Alexa Fluor secondary antibody (Molecular Probes, Eugene, OR) and TO-PRO-3 iodide (Molecular Probes) diluted 1:1000 in 3% bovine serum albumin and PBS. Cells were washed twice with PBS/bovine serum albumin and twice with PBS and then mounted on glass slides using VECTASHIELD with 4Ј,6-diamidino-2-phenylindole (Vector Laboratories, Burlingame, CA). Fluorescent images were acquired on an LSM 510 confocal microscope (Carl Zeiss, Jena, Germany) using argon (488 nm), HeNe1 (543 nm), and HeNe2 (633 nm) lasers and imaged with LSM image browser software.
In Vitro Chk2 Kinase Assays-Nuclear lysates were prepared from mock-or HpX (Tax-expressing plasmid)-transfected HEK293 cells. Lysates were incubated with 40 ng of glutathione S-transferase (GST)-Chk2 (Upstate) for 2 h on ice. GST-Chk2containing lysates were incubated at 30°C for 10 min in 1ϫ kinase buffer (20 mM Tris (pH 7.5), 10 mM MgCl 2 , 10 mM MnCl 2 , and 1 mM dithiothreitol) supplemented with 2 M unlabeled ATP and 10 Ci of [␥-32 P]ATP (Pierce). The reaction mixture was resolved by 10% SDS-PAGE, dried, and subjected to phosphorimaging using a Typhoon scanner (GE Healthcare). pcDNA4-Chk2 and pcDNA3-Tax (kindly provided by Dr. Ralph Grassmann, Institute for Clinical and Molecular Virology, Schlossgarten, Germany) or a control plasmid expressing a nonspecific protein were subjected to in vitro transcription/ translation using a rabbit reticulocyte lysate system (Promega Corp., Madison, WI). A standard 50-l reaction was performed following the recommendations of the manufacturer. 8 l of the in vitro translation product was mixed with 300 l of NETN buffer (20 mM Tris-HCl (pH 8.0), 0.1 M NaCl, 1 mM EDTA, 0.5% Nonidet P-40, and protease inhibitor mixture) for immunoprecipitation using 2 g of anti-Xpress tag antibody. Precipitates were washed twice with NETN buffer lacking protease inhibitors, followed by a final wash with 1ϫ kinase buffer. The precipitated Chk2 immune complexes were used in the kinase assays.
Chromatin Egress Assays-Cells were washed twice with cold PBS and resuspended in buffer A (10 mM HEPES (pH 7.9), 10 mM KCl, 1.5 mM MgCl 2 , 0.34 M sucrose, 10% glycerol, 1 mM dithiothreitol, 10 mM NaF, 1 mM Na 2 VO 3 , and protease inhibitor mixture). Triton X-100 was added to a final concentration of 0.1%, and the cells were incubated for 5 min on ice. Cytosolic proteins (S2) were separated from nuclei by centrifugation for 5 min at 1300 ϫ g. Nuclei were washed once with buffer A and then lysed by incubation for 30 min in buffer B (3 mM EDTA, 0.2 mM EGTA, 1 mM dithiothreitol, and protease inhibitor mixture). Insoluble chromatin was separated from soluble nuclear proteins (S3) by centrifugation for 4 min at 1700 ϫ g, washed once with buffer B, and collected by centrifugation for 1 min at 10,000 ϫ g. The final chromatin pellet was resuspended in M-PER protein lysis buffer with brief sonication. The chromatin-associated proteins (P3) were collected by centrifugation for 10 min at 12,000 ϫ g. The protein concentration in each fraction was determined by the Bradford assay as described above.
Purification of S-Chk2 and Mass Spectrometry Analysis-pTriEx4-Chk2 (coding for the fusion protein S-Chk2) was transfected into 293T cells, and three cellular fractions (cytoplasmic, nuclear soluble, and chromatin) were generated using the method described above for the chromatin egress assays. Chromatin-bound fractions were sonicated prior to performing S-Chk2 purification. S-Chk2 was purified from the lysates using the protocol recommended by Novagen. The S-Chk2 protein was purified on a standard 10% SDS-polyacrylamide gel. Protein bands were excised and cut into 1-2-mm cubes, washed three with 500 l of ultrapure water, and incubated in 100% acetonitrile for 45 min. Samples were reduced with 50 mM dithiothreitol at 56°C for 45 min and then alkylated with 55 mM iodoacetamide for 1 h at room temperature. The material was dried in a SpeedVac, rehydrated in a 12.5 ng/l modified sequencing-grade trypsin solution (Promega Corp.), and incubated in an ice bath for 40 -45 min. The excess trypsin solution was then removed and replaced with 40 -50 l of 50 mM ammonium bicarbonate and 10% acetonitrile (pH 8.0), and the mixture was incubated overnight at 37°C. Elastase digestions were performed as described for trypsin at an enzyme concentration of 15 ng/l without acetonitrile in the reaction buffer. Peptides were extracted twice with 25 l of 50% acetonitrile and 5% formic acid and dried in a SpeedVac. Digests were resuspended in 20 l of buffer containing 5% acetonitrile, 0.1% formic acid, and 0.005% heptafluorobutyric acid, and 3-6 l was loaded onto a fused silica capillary column (12 cm ϫ 0.075 mm) packed with 5-m diameter C 18 beads (The Nest Group, Inc., Southborough, MA) using an N 2 pressure vessel at 1100 p.s.i. Peptides were eluted over 55 min by applying a 0 -80% linear gradient of buffer containing 95% acetonitrile, 0.1% formic acid, and 0.005% heptafluorobutyric acid at a flow rate of 150 l/min with a pre-column flow splitter, resulting in a final flow rate of ϳ200 nl/min directly into the source. An LTQ TM linear ion trap (ThermoFinnigan, San Jose, CA) was run in an automated collection mode with an instrument method composed of a single segment and five data-dependent scan events with a full mass spectrometry scan followed by four tandem mass spectrometry scans of the highest intensity ions. The normalized collision energy was set at 35, and activation Q was 0.250 with minimum full scan signal intensity at 1 ϫ 10 5 with no minimum tandem mass spectrometry intensity specified. Dynamic exclusion was turned on utilizing a 3-min repeat count of 2 with the mass width set at m/z 1.0. Sequence analysis was performed with TurboSEQUEST TM (ThermoFinnigan) and MASCOT (Matrix Sciences, London, UK) using an indexed human subset data base of the NCBI non-redundant protein data base (www.ncbi.nlm.nih.gov/). An additional data base containing only the cDNA sequence of the Chk2 protein was created and used for mapping specific phosphopeptides.

Tax Expression Results in Binding to and Activation of Endogenous Chk2-We demonstrated previously that Tax interacts
with Chk2 by co-immunoprecipitation (26). This interaction was determined via immunoprecipitation following cotransfection of plasmids expressing both Tax and Chk2 and by co-immunoprecipitation following immunoprecipitation of endogenous Chk2. In the studies described herein, we used affinity purification of S-Tax using a system we recently described (35). In this study, the affinity isolation of Tax protein from cells resulted in a multiprotein complex containing endogenous Chk2 as a component (Fig. 1A). This complex also contained a known Chk2-interacting protein, 53BP1. The presence of both proteins was determined by gel excision and subsequent tandem mass spectrometry analysis. Neither of these proteins was associated with expression of S-GFP under conditions normalizing total protein. The presence of Chk2 in the Tax complex OCTOBER 5, 2007 • VOLUME 282 • NUMBER 40 was also confirmed by Western analysis of the same fractions (Fig. 1B).

Tax-1-induced Chk2 Oligomerization and Phosphorylation
We next examined the impact of Tax upon endogenous Chk2 and the Chk2 target protein p53. When total protein and total Chk2 were normalized, we observed an increase in the steadystate levels of Chk2 phosphorylated at Thr 68 and p53 phosphorylated at Ser 15 or Ser 20 (Fig. 1C). Thus, Tax expression results in the accumulation of "activated" Chk2 that is coincident with phosphorylation of p53 at Ser 15 and Ser 20 , both targets of Chk2 kinase.
Tax Induces Chk2 Kinase Activity In Vitro-Autophosphorylation within the transition loop (T-loop) of the kinase domain is an essential early event in the formation of a kinaseactive form of Chk2 (36,37). In an effort to directly investigate the effect of Tax on Chk2 autophosphorylation, we performed in vitro Chk2 kinase assays. An identical concentration of purified GST-Chk2 was incubated with nuclear lysates prepared from either Tax-expressing or non-Tax-expressing cells. In parallel experiments, Tax or control cell nuclear lysate was added at increasing amounts to the purified GST-Chk2 protein, and kinase assays were performed to assess autophosphorylation of Chk2. The reaction mixture was resolved by SDS-PAGE and analyzed by phosphorimaging. The incorporation of labeled ATP into the GST-Chk2 100-kDa band was measured. The data presented in Fig. 2A show that phosphorylation of Chk2 increased with the amount of nuclear lysate containing Tax. The increasing amount of control nuclear lysate reflected the background levels of Chk2 phosphorylation.
In another approach, we attempted to more directly assess the ability of Tax to induce Chk2 autophosphorylation in the absence of potential confounding factors present in nuclear extracts. Specifically, we repeated the kinase assays using the rabbit reticulocyte system to generate Tax and Chk2 proteins. The synthesized proteins were mixed in equal amounts with bovine serum albumin added to normalize for total protein under Tax-positive and Tax-negative conditions. An equivalent amount of Chk2 was then immunoprecipitated, and Chk2 immune complexes were used to perform kinase assays as described under "Experimental Procedures." The reaction mixture was resolved by SDS-PAGE and analyzed by phosphorimaging. Fig. 2B shows that the autophosphorylation activity of Chk2 increased in the presence of Tax protein. This result is consistent with a direct action of Tax on Chk2 activation.
Tax Induces Chk2 Oligomerization and Phosphorylation-It has been reported that oligomerization is a critical factor in the stabilization and activation of Chk2 (38). Therefore, we hypothesized that Tax expression, which results in increased phosphorylation of Chk2, enhances the oligomerization and increased stability of Chk2 protein complexes. To test this hypothesis, we employed a reciprocal tagging approach using molecular probes with different affinities to demonstrate homodimerization of Chk2. Our fusion constructs consisted of HA-Chk2 and Xpress-Chk2. Oligomerization of the Chk2 proteins was confirmed by immunoprecipitation with HA and immunoblotting of the precipitates with anti-Xpress antibody. We normalized for equal amounts of affinity-isolated HA-Chk2 so that the quantity of reciprocal Xpress-Chk2 was directly proportional to the amount of oligomeric Chk2. The normalization for precipitated HA-Chk2 is critical for evaluating an increase in the proportion of oligomeric Chk2.
Using this oligomerization assay, we analyzed the effect of Tax upon oligomerization of Chk2. Cells were transfected with constructs coding for HA-Chk2 and Xpress-Chk2 in the presence or absence of Tax. Whole cell lysates were subjected to immunoprecipitation targeting of the HA domain of the Chk2  A, purified GST-Chk2 was incubated with increasing amounts of nuclear lysates or nuclear lysates containing Tax protein as indicated at 4°C for 3-4 h. Kinase assays were performed as described under "Experimental Procedures." After completion of the kinase reaction, the mixture was subjected to SDS-PAGE analysis followed by phosphorimaging. The band corresponding to phosphorylated GST-Chk2 is shown. B, Chk2 kinase assays were performed using in vitro transcribed Tax and Xpress-Chk2. After completion of the kinase reaction, the mixture was subjected to SDS-PAGE analysis followed by phosphorimaging. The results shown are representative of three different independent experiments.

Tax-1-induced Chk2 Oligomerization and Phosphorylation
fusion protein with anti-HA polyclonal antibody. The immune complexes were separated by SDS-PAGE and reciprocally probed for Xpress-Chk2. The results from this experiment are presented in Fig. 3A. We observed an increase in Xpress-Chk2 in the presence of Tax protein reflective of the level of oligomeric Chk2. Thus, Tax expression results in autophosphorylation, oligomerization, and stabilization of Chk2.
We next confirmed that the reciprocal tag system faithfully reproduced the Tax-induced increase in steady-state Chk2 protein levels. Whole cell lysates were prepared, analyzed by SDS-PAGE, and subjected to Western blot analysis as described. The blots were probed for Xpress-Chk2. Fig. 3B shows that coexpression of Tax resulted in increased steady-state Chk2 protein levels. Next, we explored whether the increased protein was the phosphorylated form of Chk2. Because phosphorylation at Thr 68 is a prerequisite for the activation and oligomerization of Chk2 (38), the lysates were also probed with anti-phospho-Chk2 Thr 68 antibody. Thus, the normalization for total protein revealed a relative increase in total Chk2 as well as increased phospho-Chk2 in the presence of Tax.
Tax Inhibits DNA Damage-induced Chk2 Egress from Chromatin-As part of a normal DNA damage response, Chk2 is rapidly phosphorylated and becomes dislodged from the chromatin, so the active form is soluble in the nucleus (39). We evaluated the chromatin-associated levels of endogenous Chk2 as a qualitative measure of a normal damage response. The Chk2 egress response to DNA damage was examined in the presence or absence of Tax protein. Tax-expressing and non-Tax-expressing cells were subjected to ionizing radiation, harvested, and analyzed for the relative distribution of endogenous Chk2 in chromatin, soluble nuclear, and cytoplasmic cellular fractions. Each of the soluble extracts was normalized against the presence of known stable cellular proteins. The cytoplasmic form of Chk2 demonstrated a minor population of Chk2 phosphorylated at Thr 68 , which was dramatically increased by expression of Tax in the absence of DNA damage (Fig. 4A). Total endogenous Chk2 levels appeared to be unchanged in the cytoplasmic fraction. Ionizing radiation also induced a rapid increase in cytoplasmic Chk2 phosphorylated at Thr 68 within a background of stable total Chk2 protein. In the case of the soluble nuclear fraction, Tax alone did not induce increased levels of Thr 68 -phosphorylated Chk2, as was the case for the cytoplasmic fraction (Fig. 4B). However, Tax expression resulted in a noticeable increase in the ionizing radiation (IR)-induced levels of Thr 68 -phosphorylated Chk2. This additive effect of Tax expression persevered through the 24-h time point. Contrary to the cytoplasmic fraction, the soluble nuclear fraction of Chk2 revealed that an increased proportion of total Chk2 was Thr 68phosphorylated Chk2.
We were unable to detect Chk2 phosphorylated at Thr 68 in the chromatin, similar to results from a previous study (39). As a result of the low amounts of phospho-Chk2, egress from the chromatin fraction is generally measured by tracking changes in total Chk2. We saw a marked reduction in total Chk2 following IR (Fig. 4C), similar to the results from the previous study. In stark contrast, Tax-expressing cells displayed only slight changes in total chromatin-bound Chk2, indicating an impaired response to DNA damage. This result was reproduced three time to generate the lower panel in Fig. 4C.

Tax Expression Results in Hyperphosphorylation of the T-loop Region of Chromatin-bound
Chk2-Chk2 dimerization promotes phosphorylation of the so-called T-loop within the kinase domain of Chk2 (40). The resulting T-loop phosphorylation "exchange" confers activity to the Chk2 oligomer. This phosphorylation series is critical to the Chk2 damage response. The egress from chromatin also appears to be linked to phosphorylation, with Thr 68 -phosphorylated Chk2 being preferentially located in the soluble nuclear and cytoplasmic fractions. We thus examined the impact of Tax on phosphorylation of chromatin-bound Chk2 using direct mass spectrometry analysis.
We first demonstrated that exogenously expressed S-Chk2 localized to cytosolic, nuclear, and chromatin fractions, similar to endogenous Chk2 (Fig. 5A).The isolated Chk2 was then purified by SDS-PAGE, excised, and analyzed by tandem mass spectrometry as described under "Experimental Procedures." We were able to achieve 97% protein coverage, including all serine, threonine, and tyrosine residues. We confirmed phosphorylation at Thr 68 , Thr 383 , and Thr 387 and determined novel phosphorylation at Thr 378 , Ser 379 , Thr 389 , and Ser 517 (supplemental Fig. 1A). We were unable to confirm previous reports of phosphorylation at Ser 516 (37). Using this protocol, we then compared the changes in phosphorylation at these sites between IR-induced Chk2 and Tax-  OCTOBER 5, 2007 • VOLUME 282 • NUMBER 40 induced Chk2. The right panel in Fig.  5B shows a depiction of the regions within Chk2 that were interrogated. The left panel is a comparison of the changes in phosphorylation using manual experimental ion counting. Specifically, we determined the percent of ions representing phosphorylated peptides from total homologous peptides across five experimental runs, and this percentage was then graphed. We confirmed early studies by observing increased phosphorylation of Chk2 in nuclear and cytoplasmic fractions in response to IR. Overall, we saw the greatest amount of phosphorylation of Chk2 in all fractions for Tax-expressing cells. We noted that, upon exposure to IR or Tax, the percent of phosphorylation at Thr 383 and Thr 387 increased specific to the chromatin fraction. Additionally, when we investigated chromatin-bound Chk2 exposed to Tax expression, there were increases in the novel sites of Thr 378 , Ser 379 , and Thr 389 . The results are supportive of  10 -12 and 14 -16) of IR or in its absence (lanes 1, 5, 9, and 13). The times following IR are indicated. Also shown are the bands corresponding to phospho (P)-Chk2, Chk2, Tax, and ␣-tubulin. Cytosolic (A) and nuclear (B) fractions were both examined. C, Chk2 was expressed alone (lanes 1 and 2) or with Tax (lanes 3 and 4) in the presence of 10 grays of IR and harvested at the indicated times (upper panel). Total Chk2, Tax, and Orc-2 are indicated. The amount of chromatin-bound Chk2 following IR was determined in relation to Orc-2. This experiment was repeated three times, and the values are presented in the lower panel.  1 and 4), nuclear (lanes 2 and 5), and chromatin (lanes 3 and 6) cellular fractions. Exogenous Chk2 (lanes 4 -6) displayed subcellular distribution comparable with that of endogenous Chk2 (lanes 1-3). B, shown is a representation of the kinase domain within the Chk2 protein (right panel). The observed phosphorylation sites as identified by tandem mass spectrometry are identified. Phosphorylation at Thr 383 and Thr 387 has been described; the novel sites (Thr 378 , Ser 379 , and Thr 389 ) are presented in red. Also shown is a histogram of the estimated percent of protein phosphorylated at given residues (left panel). The relative percent of observed phosphopeptides for specific residues in Chk2 isolated from cytoplasmic (Cyto), nuclear (Nucl), and chromatin (Chrom) cellular fractions is given. The results for Chk2 from untreated cells (TaxϪ/IRϪ), cells exposed to IR (TaxϪ/IRϩ), and cells expressing Tax (Taxϩ/IRϪ) are shown.

Tax-1-induced Chk2 Oligomerization and Phosphorylation
the presence of a uniquely phosphorylated form of chromatinbound Chk2 in Tax-expressing cells.
Mapping the Chk2 Domain Responsible for Interaction with Tax-There are three well characterized domains in Chk2 protein, the serine/threonine catalytic domain, the FHA domain, and the kinase domain, each with associated Chk2 function. We wanted to identify the role of these domains in facilitating the Tax-Chk2 interaction. For this purpose, Xpress-tagged deletion mutations of Chk2 were generated and examined for the ability to bind Tax protein. Fig. 6A shows a schematic of the Chk2 deletion mutants. The Xpress-tagged mutant Chk2 proteins were coexpressed with Tax protein. The subsequent whole cell lysates were subjected to immunoprecipitation using anti-Tax polyclonal antibody, and the resulting immune complexes were probed for Xpress-Chk2. The Chk2 protein able to bind Tax was co-immunoprecipitated and subsequently detected. Fig. 6B shows the results of the Tax-Chk2 interaction mapping. We observed that full-length Chk2 and mutant Chk2 containing the kinase domain were able to bind to Tax. Deletion of the Chk2 kinase domain (as observed for Chk2-(1-221) and Chk2-FHA) resulted in loss of Tax binding ability. When the immunoprecipitation step involving Tax was performed using nonspecific IgG, no Chk2 was precipitated (data not shown). Fig. 6B (lower panel) shows the result of whole cell lysate Western blotting performed to demonstrate that all of the Xpress-Chk2 proteins were comparably expressed. On the basis of these findings, the kinase domain of Chk2 is sufficient for interaction with Tax protein.
Tax Sequesters Chk2 into Nuclear Speckles-We described previously the localization of Tax within discrete nuclear foci we termed Tax speckled structures (TSS) (41). In addition, because we had shown previously that Tax and Chk2 colocalize to these same TSS, we were interested in testing whether the Chk2 kinase domain alone retains the ability to colocalize with Tax protein. Using immunofluorescence confocal microscopy, we determined that, in the absence of Tax expression, full-length Chk2 displayed a diffuse nuclear staining pattern with faint nuclear speckling (Fig. 7, A-D). In the presence of Tax coexpression, however, Chk2 was noticeably localized within TSS with Tax (Fig.  7, E-H). Although full-length Chk2 clearly colocalized with Tax to TSS, there was still significant diffuse nuclear staining. When we expressed the kinase domain of Chk2 alone, the observed pattern of staining was similar to that for the full-length protein, displaying a diffuse nuclear staining in the absence of Tax (Fig. 7, I-L). In sharp contrast, when the Chk2 kinase domain was coexpressed with the Tax protein, the observed pattern of staining was exclusively redirected into TSS (Fig. 7, M-P). This demonstrates that Tax is able to interact with Chk2 via the kinase domain and to sequester Chk2 into nuclear TSS.

DISCUSSION
A robust DNA damage response system is essential for faithful maintenance and replication of the genome. Appropriately, genomic insults from intrinsic or exogenous factors activate a complex yet highly efficient system involving cascades of signaling molecules. This network is designed to effectively sense DNA damage, initiate DNA repair, and coordinate cellular responses such as cell cycle arrest and apoptosis (42)(43)(44)(45)(46). The network includes sensors of DNA damage (MRN (MRE11-Rad50-Nbs1) complex), adaptors/mediators (BRCA1, MDC1/ NFBD1, 53BP1), apical signal transducers (for example, ATM/ ATR), distal signal transducing kinases (Chk1/Chk2), and downstream checkpoint effectors (Cdc25 phosphatases). These checkpoints are the backbone of an efficient DNA damage response system and are considered to be discrete functional units that integrate incoming signals and that produce an effect by regulating a diverse set of enzymatic reactions depending on the type of genomic insult. Thus, cellular genomic integrity is ensured in large part by coordination of the many individual processes outlined above (47).
We (26) and others (27) have shown that de novo Tax expression results in the accumulation of cells in the G 2 /M phase of the cell cycle, a result that has been confirmed in vivo (25). This observation is believed to be a manifestation of Tax-initiated impairment of the cellular damage repair response. We demonstrated that Tax binds to and activates Chk2 (26), a major damage-induced G 2 /M checkpoint modulator. Because Chk2 facilitates at least two major DNA damage response cascades (cell cycle arrest and apoptosis), we proposed that the constitutive "activation" of Chk2 would impair the cellular response to  OCTOBER 5, 2007 • VOLUME 282 • NUMBER 40 genomic insult. Subsequently, Park et al. (48) confirmed the binding of Tax to Chk2 and agreed with our suggestion that Tax impairs Chk2 function. However, these authors observed that Tax represses the kinase activity of Chk2, based on in vitro findings, as the model for impaired function (49). This result appears to conflict with our demonstration of Tax-mediated activation of endogenous Chk2 (26). This present study has addressed this conflict and was designed to uncover the molecular mechanism of Tax-Chk2 interaction.

Tax-1-induced Chk2 Oligomerization and Phosphorylation
Chk2 is a serine/threonine kinase that, upon activation of the ATM/ATR-mediated DNA damage response, becomes phosphorylated within the serine/threonine motif present in the serine/threonine catalytic domain at the N terminus. In a typical DNA damage response, activated ATM will mediate phosphorylation at Thr 68 in the serine/threonine catalytic domain of the inactive Chk2 monomers, which then oligomerize and undergo subsequent autophosphorylation involving Thr 383 and Thr 387 within the kinase domain (50). Thus, phosphorylation at Thr 68 and subsequent oligomerization and autophosphorylation of Chk2 are essential for regulation of Chk2 activation and subsequent signal transduction and response amplification (37,38). In this study, we have demonstrated via reciprocal tagging techniques that Tax induces Chk2 oligomerization. The Tax-induced oligomeric form of Chk2 was phosphorylated in vitro, providing evidence that Tax induces Chk2 autophosphorylation, a result that is consistent with the increased oligomerization. We further confirmed the autophosphorylation by direct assessment of the phosphorylation status of the T-loop within the Chk2 kinase domain. Because phosphorylation at Thr 383 and Thr 387 has been shown to be dependent upon Chk2 kinase activity, phosphorylation at these sites is directly reflective of autophosphorylation (37).
Although the relationship between phosphorylation of Chk2 and induction of its kinase activity appears to be a direct one, the functional impact of phosphorylation of Chk2 target proteins is less clear. For instance, p53 has been reported to be phosphorylated by Chk2 at Ser 15 , Ser 18 , Ser 20 , Ser 37 , Ser 313 , Ser 314 , Ser 366 , Thr 377 , and Thr 378 (51)(52)(53)(54). In addition, Chk2 phosphorylation of Hdmx, an inhibitor of p53, induces degradation of Hdmx and subsequent stabilization of p53 (55,56). Thus, the relationship between Chk2 activation and damage response signaling is far more complicated than proposed previously. Clearly, additional regulatory features are involved such as the proximity of other cellular factors and/or spatiotemporal compartmentalization of Chk2.
In fact, the formation and persistence of protein complexes (exemplified by IR-induced foci) are directly related to the effectiveness of the cellular response to DNA damage (reviewed in Ref. 57). Indeed, a plausible theory can be entertained in which the spatiotemporal relationship of cellular factors recruited to IR-induced foci is the foundation for coordinating the many damage-induced cellular responses. Within such a paradigm, additional regulation of downstream function is supplied via the timing of the recruitment and dismissal of factors with cellular stress signals. Consequently, Chk2 has been shown recently to exist in chromatin-bound and chromatinfree fractions. It is interesting that, in response to IR-induced DNA damage, chromatin-bound Chk2 becomes phosphorylated and egresses from the chromatin (39). In practical terms, this model suggests that, even if Chk2 were activated (kinasecompetent), a failure to egress from chromatin would result in impaired function. In this study, we have shown clearly that Tax-expressing cells display an impaired Chk2 egress response to DNA damage.
To gain insight into the mechanism of retention, we examined the phosphorylation status of chromatin-bound Chk2. Compared with exposure to IR, Tax expression resulted in retention of Chk2 that was hyperphosphorylated within the kinase domain. In fact, we observed several previously unknown phosphorylation events (Thr 378 , Ser 379 , and Thr 389 ), all of which are within the T-loop alongside the known phosphorylation sites Thr 383 and Thr 387 . Phosphorylation at these novel sites is specific for Chk2 isolated from Tax-expressing cells and was not present in IR-induced Chk2. We speculate that the Tax-induced hyperphosphorylated form of Chk2 displays a impaired functional response.
One possible model for the impaired Chk2 functional response is that Tax binds to Chk2, induces oligomerization and hyperphosphorylation, and prevents mobilization of Chk2 by retention in chromatin. A precedent for a human virus employing mislocalization of a checkpoint protein does exist, and such a model has been proposed recently for human cytomegalovirus (58). This sequestration model is supported by our demonstration that Chk2 expression shifted from a diffuse nuclear to a predominantly nuclear punctuate pattern (coincident with TSS) and colocalization with Tax in Tax-expressing FIGURE 7. Tax sequesters Chk2 into nuclear speckles. Full-length Xpresstagged Chk2 (A-H) or the Xpress-tagged Chk2 kinase domain (I-P) was expressed in HEK293 cells. In addition, S-Tax-GFP was coexpressed with fulllength Xpress-tagged Chk2 (E-H) or the Xpress-tagged Chk2 kinase domain (M-P). Chk2 and the Chk2 kinase domain were detected with anti-Xpress antibody (red; C, G, K, and O). Tax expression was detected via the GFP fusion (green; F and N). Nuclei were visualized by co-staining with TO-PRO-3 iodide (blue; A, E, I, and M). Each group of three images was merged to visualize overlap (yellow/white; D, H, L, and P). Confocal microscopy images are magnified ϫ40 with a ϫ3 zoom.

Tax-1-induced Chk2 Oligomerization and Phosphorylation
cells. It is interesting that our earlier studies demonstrated that TSS colocalize with nuclear structures termed "interchromatin granules" and contain transcription factors, "repair hot spots," and 53BP1 (26,41). Thus, although it is unclear to what extent there is overlap between the normal Chk2 nuclear space and TSS, these nuclear complexes may share many of the same cellular proteins. In fact, we propose that Tax-induced sequestration of Chk2 into TSS may directly compete with damage-mediated mobilization of Chk2. The result is that a major population of Chk2 in Tax-expressing cells is not responsive to genomic insult. Thus, although Tax induces certain physical changes in Chk2 associated with activation signals, the net result is a "saturation" of the cellular Chk2-mediated response to DNA damage. This saturation effect is a direct result of reduction in the level of "responsive" Chk2 capable of mediating a normal damage response.
At the molecular level, we have shown that Tax interacts via the kinase domain of Chk2. In fact, the kinase domain alone was sufficient for binding with the Tax protein. In addition, we have shown that Tax is able to redistribute the kinase domain protein alone from a diffuse nuclear to a discrete punctuate pattern, thus completely recapitulating the Tax-mediated redistribution of Chk2. The specific interaction with the Chk2 kinase domain provides for a novel regulatory mechanism in that other reported functional interactions of Chk2 are mediated via the FHA domain. It may be that Tax, which is known to dimerize and has been shown to enhance dimerization of basis helix-loop-helix proteins (59 -61), directly enhances dimerization of the kinase domain. Alternatively, Tax may interfere with the association of Chk2 with a cellular factor required for dimerization.
In summary, we have shown that interaction of the HTLV-1 Tax oncoprotein with Chk2 results in the induction of oligomerization and autophosphorylation. The Tax-induced forms of Chk2 display a uniquely hyperphosphorylated T-loop region within the kinase domain. Tax-induced hyperphosphorylated chromatin-bound Chk2 is impaired for IR-induced chromatin egress. Our data are consistent with a model for Tax-induced genomic instability via disruption of the spatiotemporal link between Chk2 function and the cellular DNA damage response. It appears that, in Tax-expressing cells, Chk2 is capable of generating some Chk2-mediated functions such as signaling for G 2 /M accumulation, but is impaired in the generation of other Chk2-mediated functions such as response to IR damage.