p300/CREB-binding Protein Interacts with ATR and Is Required for the DNA Replication Checkpoint*

The highly related acetyltransferases, p300 and CREB-binding protein (CBP) are coactivators of signal-responsive transcriptional activation. In addition, recent evidence suggests that p300/CBP also interacts directly with complexes that mediate DNA replication and repair. In this report, we show that loss of p300/CBP in mammalian cells results in a defect in the cell cycle arrest induced by stalled DNA replication. We demonstrate that complexes containing p300/CBP and ATR can be detected in mammalian cells, and that the downstream kinase CHK1 fails to be phosphorylated in response to stalled DNA replication in cells that lack p300/CBP. These observations broaden the roles for the p300/CBP acetyltransferases to include the modulation of chromatin structure and function during DNA metabolic events as well as for transcription.

Chromosome replication is accomplished by initiating replication forks at many origins along each chromosome. Furthermore, replication origin firing is under strict temporal and spatial control throughout the cell cycle, ensuring that each DNA strand is duplicated only once per cycle (1). Cell cycle checkpoints continuously monitor the genome to prevent mitosis until DNA synthesis has been completed or until DNA aberrations have been resolved (2). There are three distinct, but partially redundant S-phase checkpoints: the replication checkpoint, the intra-S checkpoint, and the S-M checkpoint (3). The replication checkpoint occurs due to stalled replication forks and functions to prevent initiation of DNA replication from unfired origins. The intra-S phase checkpoint is activated by DNA damage during S phase and also prevents initiation from unfired origins. Finally, the S-M checkpoint ensures that the cells do not attempt to divide before the genome has been faithfully duplicated. Failure of the S-M checkpoint results in "catastrophic " mitosis of cells that have incompletely replicated DNA. The S-M phase checkpoint is thought to ensure that mitosis does not occur until the completion of S phase. In eukaryotic cells, the S-M phase checkpoint is dependent upon protein kinases that are related to the ATR/ATM family. ATR is a mammalian gene with homology to the gene mutated in the human genetic disease ataxia telangiectasia (ATM) 2 (4,5). ATR and ATM share significant functional and sequence homology with Schizosaccharomyces pombe rad3 (6), Saccharomyces cerevisiae ESR1/MEC1 (7), and Drosophila mei-41 (8). Both ATR and ATM transduce cell cycle checkpoint signals by phosphorylating a spectrum of substrate proteins including Chk1, Chk2, and p53 (9,10). ATM phosphorylates Thr-68 of Chk2 mediating its activation, whereas ATR activates Chk1 by phosphorylating Ser-345 and Ser-317 (11)(12)(13)(14). Activated Chk1 and Chk2 phosphorylate and inactivate Cdc25C, preventing activation of the mitotic kinase Cyclin B/Cdk1 (15,16).
The protein acetyltransferases p300 and CBP are transcriptional coactivators for signal transduction cascades that regulate cell cycle progression, cellular growth, differentiation, and apoptosis. Studies in mice indicate that p300 and Cbp are both required for normal development in a gene dose-dependent manner (17,18), indicating that p300 and Cbp have at least some overlapping functions. In addition to forming a physical link between activated transcription factors and the basal transcriptional machinery, p300/CBP can enhance signal-responsive transcription by acetylating chromatin-associated proteins and consequently modifying chromatin structure and function (19,20).
In addition to transcriptional activity, recent evidence suggests that p300/CBP may also interact directly with complexes that mediate chromatin metabolism. For example, p300 has been shown to bind PCNA, associate with newly synthesized DNA, and stimulate DNA synthesis in vitro (21). In addition, acetylation of Fen1, the endonuclease important for the removal of RNA primers during Okazaki fragment maturation, by p300/CBP inhibits its DNA binding and nuclease activity (22). p300 also binds and acetylates DNA polymerase b, which is involved in base excision repair (23). p300/CBP is also in a complex with and acetylates thymine DNA glycosylase, the enzyme that recognizes and repairs mispaired thymine and uracil groups (24). Further evidence that p300 plays a role during the response to DNA damage comes from the observation that acetylation of the RecQ helicase WRN by p300 facilitates the translocation of WRN protein from the nucleolus to nucleoplasmic foci (25). The WRN protein is critical for the resolution and restart of stalled DNA replication forks and after a replication block is phosphorylated by and colocalizes with ATR (26).
Taken together, these observations suggest that p300/CBP plays an important role during DNA synthesis following DNA damage or stalled replication. In this report we show that p300/ CBP associates with the replication checkpoint protein ATR in vivo, and loss of p300/CBP in mammalian cells results in a defect in the S-M phase checkpoint.
All siRNA oligos were purchased from Qiagen. Oligos were constructed targeting Lamin A/C, 5Ј-CTGGACTTCCAGAA-GAACA, and p53, 5Ј-GCAUGAACCGGAGGCCCAU. Selection of sequences for these siRNAs was based on guidelines on the Tuschl lab website.
To stably knock down the expression of ATR and ATM, we used the BLOCK-iT Inducible H1 Lentiviral RNAi System TM (Invitrogen). Gene-specific inserts were cloned into pLenti4 and pLenti6 according to the manufacturer's instructions. The lentivirus was produced in HEK 293FT cells, and the viruscontaining media was harvested for infection. The insert sequence for ATR stable siRNA expression was 5Ј-CACCGG-CGTCGTCTCAGCTCGTCTTCAAGAGAGACGAGCTGA-GACGACGCC. The insert sequence for ATM stable siRNA expressions was 5Ј-CACCGGATTTGCGTATTACTCAGTT-CAAGAGACTGAGTAATACGCAAATCC.
Cell Culture-TAT3 cells were provided by Dr. Merl Hoekstra. Monolayer cultures of HeLa SS6, HTD114, and TAT3 cells were maintained in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum, 1% penicillin and streptomycin. HEK 293FT cells were cultured in Dulbecco's modified Eagle's medium supplemented with 10% fetal bovine serum, 2 mM L-glutamine, 0.1 mM minimal essential medium non-essential amino acids, and 1% penicillin and streptomycin. Transfection of siRNA duplex oligos was carried out using Oligofectamine (Invitrogen) according to the manufacturer's protocol for adherent cells. Plasmid DNA transfections were carried out using Lipofectamine 2000 (Invitrogen) according to the manufacturer's protocol for adherent cells.
Determination of S Phase Checkpoint and G 2 Timing-DNA synthesis was assessed by BrdUrd incorporation. Control and p300 siRNA-treated HeLa cells were incubated with 20 mg/ml BrdUrd. Every 30 min for 5 h, cells were fixed, dried, and denatured in 70% formamide, 2ϫ SSC at 70°C for 3 min. Cells were dehydrated by sequentially washing with 70, 90, and 100% ethanol for 3 min and then air dried. Cells were stained using an anti-BrdUrd antibody (BD Biosciences) at a dilution of 1:2.5. The anti-mouse fluorescein-conjugated secondary antibody (Vector) was used at a dilution of 1:200. The cells were washed and stained with DAPI (12.5 mg/ml), coverslipped, and viewed under UV fluorescence on a Zeiss Axiophot fluorescent microscope.
Replication checkpoint assays were conducted by the addition of 10 mM hydroxyurea or 10 mg/ml aphidicolin to untreated or siRNA-treated HeLa cultures. The mammalian cells were fixed in 4% paraformaldehyde for 15 min at 20°C and blocked in PBS, 10% normal goat serum, 0.1% Triton, 0.1 mg/ml bovine serum albumin for Ͼ1 h at 20°C. Primary antibodies, anti-p300 RW128 monoclonal antibody (UBI, Lake Placid, NY), anti-CBP (Santa Cruz Biotechnology, Santa Cruz, CA), diluted 1:1000, or anti-phospho-histone H3 (UBI), diluted 1:100, in blocking buffer were added for 16 h at 4°C. After a washing step, cells were incubated for 2 h at 20°C in goat anti-rabbit or anti-mouse Cy3 (The Jackson Laboratories, West Grove, PA) diluted to 1:300 in blocking buffer. Plates were stained with DAPI (12.5 mg/ml), coverslipped, and viewed under UV fluorescence on a Zeiss Axiophot fluorescent microscope.
Immunofluorescence, Immunoprecipitations, and Western Analysis-For immunofluorescence, monolayer cells were fixed with 4% paraformaldehyde for 20 min followed by permeabilization with PBS, 0.1% Triton X-100 for 1 h. Cells were blocked with PBS, 10% normal goat serum, 0.1 mg/ml bovine serum albumin, 0.1% Triton X-100 for 1 h at room temperature. After blocking, cells were incubated with a 1/100 dilution of either anti-p300 antibody (Oncogene) or anti-CBP (Santa Cruz Biotechnology, Santa Cruz, CA) for 4°C overnight. Cells were washed 3 times in PBS, 0.1% Triton X-100 and incubated for 2 h at 20°C in goat anti-rabbit or anti-mouse Cy3 (The Jackson Laboratories) diluted to 1:300 in blocking buffer. Plates were stained with DAPI (12.5 mg/ml), coverslipped, and viewed under UV fluorescence on a Zeiss Axiophot fluorescent microscope., For immunoprecipitations, cell lysates with 200 -300-g total proteins were added to protein A/G beads (Santa Cruz) and incubated on ice for 30 min. Co-immunoprecipitation of p300 and FLAG-ATR was performed using a mouse monoclonal p300 antibody (Oncogene) and protein A/G beads and incubated overnight at 4°C with gentle shaking. Co-immunoprecipitation of FLAG-ATR and p300 was performed using a mouse monoclonal anti-FLAG antibody (Sigma) and protein A/G beads and incubated overnight at 4°C with gentle shaking. Co-immunoprecipitation of endogenous p300 and ATR was performed using a mouse monoclonal p300 antibody (Oncogene) and anti-mouse IgG beads (Sigma) or a rabbit polyclonal ATR antibody (Affinity Bioreagents) and protein A/G beads (Sigma), and incubated for 4 h at 4°C with gentle shaking. Beads were pelleted for 20 s at 4,000 rpm and washed two times with TENN (50 mM Tris, pH 8.0, 5 mM EDTA, 150 mM NaCl, 0.5% Nonidet P-40). Bound proteins were eluted by boiling and resolved on 8.5-10% SDS-PAGE. For Western analysis, cells were placed on ice, washed with cold PBS, scraped in RIPA buffer (50 mM Tris-Cl, pH 7.5, 150 mM NaCl, 1 mM dithiothreitol, 0.1% SDS, 0.5% sodium deoxycholate, 1% Nonidet P-40) and protease inhibitor mixture Mini-Complete (Roche Applied Science), vigorously shaken 15 min at 4°C, insoluble material was removed by centrifugation, and protein concentrations were determined using the DC protein assay (Bio-Rad). Cell lysates (100 mg) were separated by SDS-PAGE, and transferred to polyvinylidene difluoride membrane. Blots were blocked in Tris-buffered saline, 5% milk, 0.1% Tween 20 and incubated with primary antibodies: anti-p300/CBP (number NA46; Oncogene, San Diego, CA), anti-p300 RW128 monoclonal antibody (UBI, Lake Placid, NY), anti-CBP, anti-Chk1, anti-Lamin A/C, anti-p53, and anti-PCNA (Santa Cruz Biotechnology), anti-ATM (GeneTex Inc., San Antonio, TX), anti-ATR (Affinity Bioreagents Inc., Golden, CO), anti-phospho-Chk1 (Cell Signaling Technology, Inc., Beverly, MA), anti-␣-tubulin monoclonal antibody and anti-FLAG M2 (Sigma), and alkaline phosphatase-conjugated anti-mouse or anti-rabbit antibodies (Santa Cruz Biotechnology) with signal development using ECF (Pierce).
In Vitro Translations and Pulldowns-Glutathione S-transferase (GST) fusion proteins were prepared from bacteria by sonication in PBS containing a protease inhibitor mixture (Sigma). Extracts were affinity purified by using glutathioneagarose beads (Sigma), eluted with PBS containing 10 mM glutathione, and concentrated by a Centricon (Millipore). Protein concentrations were determined by the DC protein assay kit (Bio-Rad). In vitro translated [ 35 S]methionine (New England Nuclear)-labeled ATR and its mutants were prepared using Promega TNT kit. ATR and its mutants were incubated with GST affinity-purified fusion p300 proteins for 2 h at 4°C with constant agitation. After extensive washes with TENN buffer and RIPA buffer, bound proteins were eluted by boiling in SDS sample buffer and resolved by SDS-PAGE. Radiolabeled proteins were visualized by autoradiography.

RESULTS
We previously used siRNA to deplete p300 from human tissue culture cells (27). Interestingly, 72-120 h following p300 and CBP depletion, HeLa cells begin to exhibit aberrant mitotic structures (data not shown). Similarly, Drosophila Kc cells treated with siRNA directed against dCBP begin to die after 48 h of treatment, and appear to experience mitotic catastrophe (28). Because mitotic catastrophe can result from a defect in the DNA replication checkpoint, we determined whether loss of p300/CBP function could similarly result in a deficiency in the cell cycle delay that can be induced by stalled DNA replication. For this analysis we depleted HeLa cells of p300/CBP using siRNA. Cells were treated with siRNAs for 3 days and subjected to Western blot analysis with antibodies that can distinguish between CBP and p300. Fig. 1A shows that oligo 3 eliminates both p300 and CBP protein, without affecting expression of ATR. Western blot analysis of a time course of treatment with oligo 3 indicated that p300/CBP expression, using an antibody that detects both proteins, was reduced at all time points analyzed (Fig. 1B). Furthermore, depletion of p300/CBP from HeLa cells did not affect expression of the checkpoint proteins ATR, CHK1, or a control protein Lamin A/C (Fig. 1B). To determine whether the cells depleted of p300/CBP continue in the cell cycle, we assayed the frequency of S phase cells using BrdUrd incorporation, and the frequency of mitotic cells using staining with the anti-phospho-(Ser-10) H3 antibody. Fig. 1C shows that cells depleted of p300/CBP continue to cycle for at least 96 h following siRNA treatment without a significant drop in the frequency of cells in S phase or mitosis. In addition, to determine whether loss of p300/CBP affected the time between S phase and M phase, we measured the length of G 2 in cells depleted of p300/CBP. For this analysis, we combined phospho-H3 staining with anti-BrdUrd staining at increasing times following the addition of BrdUrd. This analysis indicated that phospho-H3 and BrdUrd double positive cells were detected with similar kinetics following the addition of BrdUrd (Fig. 1D), indicating that depletion of p300/CBP did not alter the length of time that cells spend in the G 2 phase of the cell cycle. However, mitotic catastrophes were observed in cells depleted of p300/CBP at the later time points (72-120 h; data not shown). Analyzing the cell cycle arrest induced by stalled DNA replication indicated that HeLa cells treated for 48 and 72 h of p300/ CBP siRNA completely lacked the hydroxyurea-induced cell cycle arrest (Fig. 1E). In contrast, mock-treated cells retained the hydroxyurea-induced cell cycle arrest. A similar result was obtained with aphidicolin, a DNA polymerase inhibitor, indicating that the defect in the S-M phase checkpoint is not specific to hydroxyurea-induced stalled replication (Fig. 1F). Similar results were obtained with the human fibrosarcoma cell line HTD114 (supplemental Fig. S1). These results indicate that loss of p300/CBP results in a defect in the cell cycle arrest induced by stalled DNA replication.
To determine whether the S-M phase checkpoint defect observed in cells depleted of CBP and p300 is specific to one or both, we carried out a series of experiments designed to knockout CBP or p300 individually, or to replace CBP or p300 in the siRNA (oligo 3)-treated cells. To determine whether depletion of CBP or p300 would result in a defect in the cell cycle arrest induced by stalled replication, we used transient transfection of small hairpin RNA (shRNA) constructs expressing sequences that are specific to either p300 or CBP. To demonstrate the specificity of these hairpin vectors, we co-transfected a GFP expression vector to identify the transfected cells and stained the cultures with antibodies specific for p300 or CBP. Fig. 2A shows that the CBP vector depletes cells of CBP but not p300, and that the p300 vector depletes cells of p300 without affecting CBP. Scoring the GFP positive cells indicated that depletion of CBP or p300 alone does not result in a defect in the hydroxyurea-induced cell cycle arrest (Fig. 2B). However, cotransfection of both CBP and p300 shRNA vectors resulted in a defect in the hydroxyurea-induced cell cycle arrest, indicating that loss of both proteins is necessary to eliminate this cell cycle arrest.
Next, to determine whether forced expression of CBP could rescue the checkpoint defect in p300/CBP-depleted cells, we introduced a murine Cbp expression vector, by transient transfection, at the same time we introduced the siRNA against p300/CBP. In addition, to replace p300 in cells depleted of p300/CBP, we generated a pseudo-wild type p300 mutant (pwp300). This p300 mutant contains 3 nucleotide changes at the wobble positions within the oligo 3 sequence. These p300/CBP and the Replication Checkpoint FIGURE 1. A, p300/CBP is required for the cell cycle arrest induced by hydroxyurea or Aph treatment in mammalian cells. HeLa cells were exposed to siRNAs against four differentregionsofhumanp300(24)(lanes2-5).Controlsconsistedofmocktreatedcells(lane1)andcellstreatedwithsiRNAsdirectedagainstLaminA/C(lane6).Westernblots werepreparedfollowing72hoftreatment,andprobedwithantibodiesspecificforp300,CBP,andATR.p300oligo3(lane 4) resulted in cells depleted for both p300 and CBP. B, time course of p300 siRNA treatment. HeLa cells were exposed to p300 siRNA (oligo 3) for 0 (NT), 48, 72, 96, and 120 h. Western blots were prepared and probed with antibodies against p300/CBP, ATR, CHK1, and Lamin A/C. C, HeLa cells depleted for p300/CBP continue in the cell cycle. Cells were exposed to the p300 siRNA for the indicated times and assayed for BrdUrd incorporation (S-phase) and phospho-H3 (M-phase). D, G 2 timing is not altered in p300/CBP-depleted cells. HeLa cells were treated with oligo 3 (siRNA) for 48 h. BrdUrd was added to the media at time 0 and cells were fixed and assayed for BrdUrd incorporation and phosphorylation of H3 (P-H3) at the indicated time points. The frequency of BrdUrd and P-H3 positive cells, as well as BrdUrd plus P-H3 (Doubleϩ) cells was determined at the indicated times. E and F, HeLa cells depleted of p300/CBP are defective in the cell cycle arrest induced by stalled replication. The cells were stained with DAPI and the P-H3 antibody to determine the frequency of mitotic cells. The graphs show the results from three independent experiments. Cells treated with p300 siRNA for 48 and 72 h were exposed to hydroxyurea for the indicated times, and frequency of mitotic cells from the mock treated cells in the absence of hydroxyurea ranged from 6 to 9% and was set at 100%. Cells treated with p300 siRNA for 48 h (blue) or untreated (black) were exposed to Aph for the indicated times and the percent of mitotic cells is indicated. MARCH 30, 2007 • VOLUME 282 • NUMBER 13 changes do not alter the amino acid sequence but make the mRNA resistant to oligo 3. Fig. 2C shows Western blot analysis of cells transfected with oligo 3 plus expression vectors for murine Cbp, wild type p300, or pwp300. This analysis indicated that murine Cbp and pwp300 are resistant to oligo 3, but wild type p300 is sensitive to this siRNA. For the analysis of the hydroxyurea to induce cell cycle arrest in cells that received either Cbp or pwp300 we identified the transfected cells by co-transfection with a GFP expression vector. Scoring the GFP positive cells, and therefore the Cbp or pwp300 expressing cells,

. p300 and CBP are redundant in the hydroxyurea-induced cell cycle arrest. A, transient transfection of HeLa cells with a GFP expression vector
and vectors expressing small hairpin RNAs directed against p300 (pHp300) or CBP (pHCBP). Transfected cells were identified by GFP expression, and endogenous CBP and p300 expression was detected using immunofluoresence with specific antibodies. Nuclei were identified using DAPI. Arrows mark GFP positive cells. B, cells depleted of both p300 and CBP lack the hydroxyurea-induced cell cycle arrest. HeLa cells were cotransfected as in A, treated with hydroxyurea for 4 and 8 h, and the GFP positive cells were assayed for mitotic cells using the phospho-H3 antibody. The graph represents the average of three independent transfections and error bars represent mean Ϯ S.E. C and D, forced expression of either p300 or CBP can rescue the hydroxyurea-induced cell cycle arrest in cells depleted of p300 and CBP. C, HeLa cells were transfected simultaneously with oligo 3 plus expression vectors for FLAG-tagged wild-type human p300 (flag-p300), pseudo-wild-type p300 (flag-pwp300), or murine CBP (flag-CBP). A GFP expression vector was included as a transfection control. Western blots were prepared after 48 h and probed with FLAG, GFP, and tubulin antibodies. D, cells were transfected as in C and treated with hydroxyurea for 0 (Ϫ) or 8 h (ϩ). Cells transfected with empty pcDNA3 vector (Vector) served as transfection control. The graph shows the results from three independent experiments.

p300/CBP and the Replication Checkpoint
indicated that forced expression of either Cbp or pwp300 restored the hydroxyurea-induced cell cycle arrest (Fig. 2D). In addition, this analysis demonstrated that the loss of the replication checkpoint in oligo 3-treated cells is specific and is the result of depletion of p300 and CBP.
We have demonstrated that p300/CBP is required for replication checkpoint function in mammalian cells. Although ATR has been shown to be required for replication checkpoint function in Caenorhabditis elegans and Drosophila (8,29,30), surprisingly, an Atr knock-out did not effect the replication checkpoint function in mouse embryo fibroblasts exposed to hydroxyurea or aphidicolin, nor did Atr/Atm or Atr/p53 double knock-outs (31). Nonetheless, following a block in DNA replication, p300 acetylates p53 resulting in stabilization and activation of p53 (32). Furthermore, p53 has been shown to play a role in the DNA replication checkpoint in mouse embryo fibroblasts (33). Because it has been shown in mouse embryo fibroblasts that neither an Atr/Atm double knock-out nor an Atr/p53 double knock-out interferes with replication checkpoint integrity, we wished to determine whether an ATR, ATM, and p53 triple knockdown is a phenocopy of a p300/CBP knockdown. ATR and ATM have recently been stably knocked down in HeLa cells using ATR-and ATM-specific Lentivirus-expressed shRNAs (34,35). We constructed lentiviruses expressing ATR-and ATM-specific shRNAs, and after transduction of HeLa cells, individual clones were assayed for ATR and ATM levels by Western blot. Fig. 3A shows that in three separate clones, clone 1, clone 3, and clone 19, ATR levels were significantly reduced relative to an untransduced control. The same three clones showed nearly complete ablation of detectable ATM levels relative to an untransduced control, appearing similar to the ATM null cells TAT3 (Fig. 3B). Having stably knocked down ATR and ATM in three separate clones, clone 11 was transfected with siRNA oligos directed against p53. 48 h post-transfection, p53 levels were significantly reduced relative to untransfected clone 11 cells (Fig. 3C). Analyzing the cell cycle arrest induced by stalled DNA replication indicated that cells with an ATR, ATM, and p53 triple knockdown had a compromised hydroxyurea-induced cell cycle arrest (Fig. 3D). In contrast, control HeLa cells retained the hydroxyurea-induced cell cycle arrest. These results indicate that in HeLa cells, ATR, ATM, and p53 have some redundant functions in the replication checkpoint and that simultaneous knockdown of ATR, ATM, and p53 results in a phenotype similar to p300/CBP knockdown. In addition, ATM is also activated in response to replication blocks (36), and following knockdown of p300/CBP, HeLa cells exhibit a compromised G 2 /M DNA damage checkpoint (data not shown). This data suggests that several layers of redundancy may function in the mammalian DNA structure checkpoints and that are all dependent on p300/CBP functions.
To begin to determine whether p300/CBP plays a direct role during the replication checkpoint, we tested whether a p300-ATR interaction could be detected in vitro and in vivo. First, we used co-immunoprecipitation from cells transfected with p300 and ATR. Fig. 4A shows that full-length ATR, and an ATR deletion mutant ATR- , are able to coimmunoprecipitate p300. In addition, using an anti-p300 antibody, we detected coimmunoprecipitation of ATR and ATR-(2-1127) (Fig. 4B).
Because the kinase domain of ATR resides in the C terminus and is deleted in the ATR-(2-1127) mutant, these observations indicate that the kinase domain of ATR is not required for the p300 interaction. We next used a GST pull-down assay to determine which domain of ATR interacts with which domain of p300. For this analysis, we assayed a series of GST-p300 fusion constructs (Fig. 4C) for interactions with in vitro translated ATR and ATR mutant proteins. Fig. 4D shows that full-length ATR interacts with the fusion proteins containing amino acids 19 -596 and 963-1302, which contain the first and second Cys/ His domains of p300, respectively (19). In addition, the C-terminal truncation mutant, ATR-(2-1127), interacts with the same two GST-p300 proteins, indicating that the kinase domain of ATR is dispensable for this interaction. Furthermore, a second C-terminal truncation mutant, ATR-(2-264), containing the NH 2 -terminal 264 amino acids, interacts only with the GST-(963-1302) protein. These observations suggest that two different NH 2 -terminal domains of ATR interact with two different domains of p300. Finally, to determine whether a p300-ATR interaction could be detected with endogenous proteins, and to determine whether stalled DNA replication affected the p300-ATR interaction, we analyzed the p300-ATR interaction using coimmunoprecipitation of proteins from cells Several individual clones were isolated, and 2 weeks later extracts were harvested for Western blots. Three representative clones, clone 1, clone 11, and clone 19, each showed considerable decreases in ATR levels relative to control HeLa cells, whereas showing no changes in the levels of a control protein, PCNA. B, siRNA mediated p53 down-regulation in clone 11. Clone 11 was transfected with an siRNA oligo specific for p53. Extracts were harvested and Western blots were prepared 72 h post-transfection, and probed with antibodies specific for p53. C, extracts from clones 1, 11, and 19 also contained no detectable levels of ATM relative to control HeLa cells. The ATM null cell line TAT3 also contained no detectable levels of ATM. D, down-regulation of ATR, ATM, and p53 in clone 11 compromises the replication checkpoint. Clone 11 cells were transfected with an siRNA oligo specific for p53. The cells were stained with DAPI and the P-H3 antibody to determine the frequency of mitotic cells. The graphs show the results from three independent experiments. Untransfected HeLa cells (left series) and clone 11 cells treated with p53 siRNA for 72 h (right series) were exposed to 10 mM hydroxyurea for 0, 2, 4, and 8 h. MARCH 30, 2007 • VOLUME 282 • NUMBER 13 treated with hydroxyurea. Fig. 4E shows that complexes containing p300 and ATR could be detected using immunoprecipitation with an anti-p300 antibody. In addition, we detected an increase in the amount of coimmunoprecipitated ATR when the cells were exposed to hydroxyurea (Fig. 4E). These observations indicate that complexes containing ATR and p300 can be detected in vivo, and suggest that complexes containing ATR and p300 increase in response to stalled DNA replication.

p300/CBP and the Replication Checkpoint
One critical target of ATR phosphorylation is the downstream kinase CHK1. When the DNA replication checkpoint is activated by stalled DNA replication, ATR phosphorylates CHK1 on serine 345 (12). Therefore, to determine whether loss of p300/CBP interferes with activation of ATR kinase activity in response to hydroxyurea, we assayed phosphorylation of CHK1 with an antibody against phospho-Ser-345. As controls we used mock siRNA-treated cells exposed to hydroxyurea in the presence or absence of caffeine. Caffeine is an inhibitor of ATR kinase activity and consequently is an inhibitor of the DNA replication checkpoint (37)(38)(39). Fig. 4F shows that CHK1 is phosphorylated on Ser-345 following 2 and 4 h of hydroxyurea exposure in mock treated cells. In contrast, cells treated with p300/CBP siRNA showed very little CHK1 phosphorylation at either time point. As expected, caffeine interfered with the hydroxyurea-induced phosphorylation of CHK1. Finally, depletion of p300/CBP does not interfere with ATR or CHK1 expression. Taken together, these observations indicate that complexes containing p300/CBP and ATR can be detected in vivo and that p300/CBP is required for phosphorylation of CHK1 in response to stalled DNA replication.

DISCUSSION
The regulation of chromatin structure and function is essential for proper gene transcription, DNA . p300/CBP interacts with ATR and is required for activation of ATR in the presence of hydroxyurea. A and B, coimmunoprecipitation of transfected ATR and p300. A, cells were transfected with expression vectors for p300 and FLAG-tagged ATR or an ATR deletion mutant ATR-(1-1127). Cell extracts were immunoprecipitated with a FLAG antibody (IP/Flag ATR), and Western blots were prepared and probed with antibodies (IB/AB) against FLAG and p300/CBP. Cell extracts (1/10 of input) were included on the blots to serve as loading controls. B, cells were transfected as in A and cell extracts were subjected to immunoprecipitations with an antibody against p300/CBP (IP/p300/CBP), Western blots were prepared and probed with antibodies (IB/AB) to FLAG and p300/CBP. Cell extracts (1/10 of input) were included on the blots to serve as loading controls. C and D, two different NH 2 -terminal domains of ATR interact with two different domains of p300. A series of GST-p300 fusion proteins (C) were used in GST pull-down assays with in vitro translated 35 S-labeled full-length ATR or two NH 2 -terminal truncated ATR mutants. Input and GST alone were included as controls. E, coimmunoprecipitation of endogenous ATR and p300/CBP. Cells were exposed to hydroxyurea for 0, 4, 19, and 24 h. Cell extracts were subjected to immunoprecipitation with the p300/CBP antibody and precipitated proteins were processed for Western blot analysis using antibodies against ATR, p300/CBP, and tubulin. Cell extracts (1/10 of input) were included on the blots to serve as loading controls. F, cells depleted of p300/CBP show a defect in the phosphorylation of CHK1 in response to hydroxyurea. Cells exposed to oligo 3, or mock treated cells, for 48 h were incubated with hydroxyurea for 0, 2, or 4 h. Caffeine treatment (5 mM) was included during the hydroxyurea time course on one set of mock treated cells. Cell extracts were prepared and subjected to Western blot analysis using antibodies against phosphorylated CHK1 (S345), CHK1, p300/CBP, ATR, and tubulin.

p300/CBP and the Replication Checkpoint
replication, recombination, and repair. Acetylation of histones and other chromatin proteins by a variety of acetyltransferases is one mechanism whereby cells across phylogenetic lineages modulate a large number of cellular functions. Mutations in the p300/CBP family of acetyltransferases cause lethal developmental defects in mouse, humans, and Drosophila (19). How acetylation of chromatin proteins modulates chromosome structure and function to affect these cellular functions is still poorly understood.
Most of the accumulated data implies that p300/CBP affects the cell cycle indirectly through its ability to stimulate the transcription of genes encoding proteins required for replication and other cell cycle regulatory processes (19). However, recent reports provide evidence that p300/CBP can interact directly with and acetylate DNA metabolic enzymes (21,23,24). In this report we describe evidence suggesting that p300/CBP may also mediate cell cycle arrest in response to stalled DNA replication through interactions with ATR.
When HeLa cells are depleted of both p300 and CBP, a defect in the G 2 -M checkpoint and a marked mitotic catastrophe phenotype occurs. Similarly, depletion of ATR in human cells results in chromosome fragmentation during mitosis (40). In addition, disruption of the Atr gene in mice leads to chromosome fragmentation and caspase-dependent apoptosis leading to early embryonic lethality (41). Although the mitotic catastrophe and chromosome shredding that results from loss of p300/CBP function results in cell death, we have not determined whether or not these cells also experience apoptosis. However, unlike the Mei-41 pathway in flies and atl-1 in C. elegans, mammalian ATR does not appear to be a required component in hydroxyurea-mediated cell cycle arrest. ATR-depleted, as well as ATR/ATM-and ATR/ p53-depleted, murine embryonic fibroblasts exhibit an intact hydroxyurea-dependent cell cycle arrest (31). We have demonstrated, however, that simultaneous knockdown of ATR, ATM, and p53 in HeLa cells results in a compromised S-M checkpoint in response to hydroxyurea exposure suggesting that these three proteins have some degree of redundancy in this checkpoint and that each may require p300/ CBP. We propose a model in which p300/CBP functions by regulating the activities of multiple proteins at both normal and paused replication forks (Fig. 5). p300/CBP has been shown to be physically associated with PCNA in vivo and to positively regulate PCNA activity at replication forks by acetylation (21,42). ATR has been shown to be chromatin associated in S phase in undamaged human cells (43,44). In chromatin binding assays using Xenopus egg extracts, ATR was shown to associate with chromatin after initiation of DNA replication, and additional levels of ATR became chromatin associated by blocking DNA replication with the addition of aphidicholin. This association was disrupted by actinomycin D, an inhibitor of RNA primase, suggesting that at least a fraction of ATR may be associated with normal replication forks in S phase but that the level of chromatin-associated ATR increases with a replication block (45). Because we have shown that at least a fraction p300 is constitutively associated with ATR, even in the absence of a replication block and this association is enhanced when DNA replica-tion is blocked, both p300 and ATR may associate with and function at normal and stalled replication forks. The recruitment of ATR to stalled replication forks requires the single strand-binding protein RPA that loads onto regions of single-stranded DNA (46). It will be interesting to determine whether the p300-ATR interaction is required for either the initiation or maintenance of ATR recruitment to stalled replication forks. After replication block, a number of other proteins are recruited to the stalled replication fork including Rad17, the Rad9-Rad1-Hus1 complex, the RecQ helicases WRN and BLM, Rad51, BRCA1, and p53 (25,26,(47)(48)(49)(50)(51). The RecQ helicase WRN colocalizes with and is phosphorylated by ATR following replication arrest, and is required for the correct recovery of cells after replication arrest (26,49). The DNA damage-induced translocation of WRN from the nucleolus to nucleoplasmic foci has been shown to be regulated by p300/CBP (25). Following replication arrest, WRN has also been shown to form a complex with Fen1, an endonuclease involved in the base excision repair pathway (52). Furthermore, p300/CBP binds to and acetylates Fen1 as well as several other components of the FIGURE 5. p300/CBP regulates the activities of multiple proteins at both normal and paused replication forks. A, normal, unstressed, replication is shown. p300/CBP directly binds to and acetylates PCNA, positively regulating its activity. ATR associates with a normal replication fork by binding to both single-stranded DNA-bound RPA and p300/CBP. B, following a replication block, increasing amounts of ATR associated with single-stranded DNAbound RPA and a number of other proteins are recruited including the Rad9-Rad1-Hus1 complex, the RecQ helicases WRN and BLM, Rad51, BRCA1, and p53. p300 positively regulates WRN recruitment through acetylation and p300 also acetylates p53. In addition, p300 negatively regulates members of the base excision repair pathway by acetylating Fen1, DNA polymerase ␤, and the DNA glycolase NEIL2.

p300/CBP and the Replication Checkpoint
base excision repair pathway, and has been proposed to either play a role in down-regulation of this pathway or preventing its inappropriate activation at DNA lesions that do not require the base excision repair pathway (3,23,53). Finally, p300 associates with and activates p53 in response to stalled replication resulting in transcriptional activation of p21 and gadd45 (32,36,54).
In addition, ATM must be acetylated by the Tip60 histone acetyltransferase to be activated following ionizing radiation (55). Because, ATM has also recently been shown to be activated by stalled replication (36), it will be interesting to determine whether ATR activation also involves acetylation. Studies from yeast indicate that acetyltransferases are critical components of pathways that maintain chromatin structure. For example, genetic studies of the GNAT and MYST acetyltransferases demonstrate that histone tails are differentially acetylated to bring about activation of transcription, DNA repair, and cell proliferation. Yeast mutant for both the GNAT acetyltransferases Gcn5 and the MYST acetyltransferase Sas3 have a global reduction in histone H3 acetylation, are nonviable, and blocked at the G 2 /M transition (56). In addition, mutation in the MYST acetyltransferase Esa1 results in a global reduction of histone H4 acetylation, cell lethality, and a defect in the Rad9 DNA repair checkpoint (57,58). Taken together, these results support a model in which acetyltransferases in both yeast and higher eukaryotes modulate chromatin structure in response to signal transduction pathways that regulate transcription, DNA repair, recombination, and replication. The fact that loss of p300/ CBP function results in checkpoint defects and mitotic collapse suggests that these acetyltransferases are critically important for the maintenance of genome integrity.