Down-Regulation of Histone H2B by DNA-Dependent Protein Kinase in Response to DNA Damage Through Modulation of Octamer Transcription Factor 11

[CANCER RESEARCH 63, 7197–7205, November 1, 2003] Down-Regulation of Histone H2B by DNA-Dependent Protein Kinase in Response to DNA Damage through Modulation of Octamer Transcription Factor 11

Caroline Schild-Poulter,2 Amy Shih, Nicholas C. Yarymowich, and Robert J. G. Hache´2 Departments of Medicine [C. S-P., A. S., N. C. Y., R. J. G. H.], Cellular and Molecular Medicine [C. S-P.], and Biochemistry, Microbiology and Immunology [R. J. G. H.], University of Ottawa, The Ottawa Health Research Institute, Ottawa, Ontario K1Y 4E9, Canada

ABSTRACT catalytic activity is required for DNA repair (3, 9, 13–15). DNA-PK has been shown to phosphorylate in vitro other participants of the Cells respond to double-stranded DNA breaks (DSBs) by pausing cell nonhomologous end-joining complex, including XRCC4 and Artemis cycle progression to allow the repair machinery to restore genomic integ- rity. DNA-dependent protein kinase (DNA-PK), comprising a large cata- (14, 16, 17), but the roles of these phosphorylations remain unclear. In addition to participating directly in DNA joining, DNA-PK may lytic subunit (DNA-PKcs) and the Ku antigen regulatory subunit (Ku70/ Ku80), is activated in response to DSBs and is required for DNA repair also regulate DNA-damage-response-signaling pathways. Establish- through the nonhomologous end-joining pathway. Here we provide evi- ing a role for DNA-PK in DNA-damage response signaling has been dence that DNA-PK participates in altering specific gene expression in complicated by the overlapping specificity of DNA-PK and ATM, a response to DNA damage by modulating the stability and transcriptional closely related kinase (1, 18–20). After a flurry of reports ascribing regulatory potential of the essential transcription factor octamer tran- DNA damage-dependent signaling to one or the other kinase, it is now scription factor 1 (Oct-1). Histone H2B and U2 RNA, whose expression established that ATM is a key regulator of checkpoint responses and are highly dependent on Oct-1, were strongly decreased in response to ionizing radiation in a DNA-PK-dependent manner, and Oct-1-dependent an activator of the homologous recombination repair pathway (1, 21, reporter gene transcription was repressed. Furthermore, Oct-1 phospho- 22). In contrast, DNA-PK appears to be important for the effect of p53 rylation in response to ionizing radiation increased in a DNA-PK-depend- in mediating apoptosis (23, 24). Several studies have also implicated ent manner. Paradoxically, down-regulation of transactivation correlated DNA-PK in the modification of RPA2, nuclear factor-␬B, and H2AX with the rapid DNA-PK-dependent stabilization of Oct-1. Stabilization of after IR (25–27). Recently, we also provided evidence that auto- Oct-1 was dependent on the NH -terminal region of Oct-1, which contains 2 inactivation of DNA-PK is essential for cell survival at a step after a transcriptional activation domain and which was phosphorylated by DNA repair (28). DNA-PK in vitro. These results suggest a mechanism for the regulation of Oct-1 in response to DNA damage through specific phosphorylation IR-induced DNA damage results in profound changes in gene expression as the cells suspend progression through the cell cycle and within the NH2-terminal transcriptional regulatory domain. repair their DNA. Histone gene transcription, for example, has been INTRODUCTION reported to be down-regulated (29–31). There are many reports in the literature of DNA-PK acting to regulate specific gene transcription DSBs3 induced by IR or other DNA damaging agents activate a (32–37). For example, we have demonstrated that recruitment of signal transduction cascade leading to the assembly of DNA repair DNA-PK to the mouse mammary tumor virus promoter represses the complexes, cell cycle arrest and changes in specific gene transcrip- induction of viral expression in response to steroid hormones (5). tion. The DNA-PK complex is an essential component of the DBS However, whether DNA-PK directly regulates specific gene transcrip- repair pathway. DNA-PK is a Ser/Thr kinase composed of Ku, a tion in response to DNA damage has not been established. heterodimer of two subunits (Ku70/Ku80) and a large catalytic sub- Recently, we showed that Ku antigen exhibits a specific interaction unit, DNA-PKcs, member of the phosphatidylinositol-3-related kinase with the homeodomain of Oct-1 in a manner that strongly promotes family (1–3). Ku is the DNA-binding subunit of the complex that phosphorylation by DNA-PK from DNA ends (38). Ku-dependent recruits DNA-PKcs to DNA ends and some DNA sequences to acti- phosphorylation of Oct-1 by DNA-PK may be representative of a vate the catalytic activity of the kinase (3–6). broader effect because Ku was also determined to interact similarly DNA-PK is required for nonhomologous end-joining, which is the with several other homeodomain proteins, including HOXC4 and predominant repair process in cycling mammalian cells (7–9). Mice Dlx2. deficient in any component of this complex are radiosensitive, immu- Oct-1, a member of the POU homeodomain family, is a ubiquitous nodeficient, and develop tumors (10, 11). Cells derived from these and essential transcription factor that regulates the expression of mice show an increased number of chromosomal aberrations, includ- several genes essential to cell viability (39–41). For example, Oct-1 ing chromosome fusions and telomeric fusions, which have suggested has been shown to be required for the S-phase-dependent activation of a “genome caretaker” role for the DNA-PK complex (9–12). Ku transcription of histone H2B (42) and has been shown to regulate the recognizes the broken DNA ends and recruits DNA-PK , whose cs expression of several small nuclear RNA genes, including the RNA polymerase II-dependent gene U2 and the RNA polymerase III- Received 2/11/03; revised 8/14/03; accepted 8/18/03. The costs of publication of this article were defrayed in part by the payment of page dependent gene U6 (43–47). Oct-1 has also been implicated in the charges. This article must therefore be hereby marked advertisement in accordance with regulation of DNA replication from viral and from mammalian origins 18 U.S.C. Section 1734 solely to indicate this fact. 1 Supported by an operating grant from the Cancer Research Society Inc. (CRS) to of replication (48, 49). R. J. G. H. and C. S-P. R. J. G. H. is an Investigator of the Canadian Institutes of Health One of the transcriptional responses to DSBs that has recently been Research. reported is the down-regulation of histone H2B transcription (50). In 2 To whom requests for reprints should be addressed, at The Ottawa Health Research Institute, 725 Parkdale Avenue, Ottawa, Ontario K1Y 4E9, Canada, Phone: (613) 761- the present study we show that U2 expression is also down-regulated 5142; Fax: (613) 761-5036; E-mail: [email protected] [C. S-P.] or [email protected] in response to DSBs and that this down-regulation of histone H2B and [R. J. G. H.]. 3 The abbreviations used are: DSB, DNA double-stranded break; IR, ionizing radia- U2 expression is dependent on DNA-PK. This effect appeared to be tion; DNA-PK, DNA-dependent protein kinase; ATM, ataxia telangiectasia-mutated; mediated through a DNA-PK-dependent down-regulation of the trans- Oct-1, octamer transcription factor 1; GST, glutathione S-transferase; aa, amino acid; CAT, chloramphenicol acetyltransferase; CMV, cytomegalovirus; ␤-Gal, ␤-galactosidase; activation potential of Oct-1 in a manner that paradoxically also C/EBP, CCAAT enhancer-binding protein. results in stabilization of the protein. These results establish a role for 7197

DNA-PK in regulating specific gene transcription in response to Santa Cruz, CA), Ku70 (N3H10; NeoMarkers, Fremont, CA), ␤-actin (AC-15; DSBs. Sigma Chemical Co., St. Louis, MO), or HA (HA11; Covance, Princeton, NJ). The blots were developed with Enhanced Luminol Reagent (Western Light- ning; NEN Perkin-Elmer Life Sciences, Boston, MA). Protein levels were MATERIALS AND METHODS quantified by ImageQuant software (Molecular Dynamics, Sunnyvale, CA). Values obtained with HeLa cells were from eight samples analyzed in two Plasmids. pGEX 3X-Oct-1 POU has been described previously (38). separate experiments. Oct-1 protein levels measured in V79/V15B and M059 pGEX 3X-Oct-1 was generated by cloning the full-length Oct-1 cDNA isolated K/J were obtained in a minimum of three separate experiments. Oct-1 levels were analyzed in extracts of unirradiated cells (control) or cells taken at from pCGNOct-1 (51) in the EcoRI site of pGEX 3X. GST-Oct-1 NH2- terminal was produced by inserting an Oct-1 fragment (aa 1–268) in pGEX 2T, various time points between 1 and 8 h after irradiation and calculated relative whereas the GST-Oct-1 COOH terminus resulted from the cloning of Oct-1 aa to their respective endogenous actin levels (ϮSE). 414–743 in pGEX 3X. Deletion mutants of Oct-1 were generated from the Northern Blots. RNA was extracted with the RNeasy purification kit pCGNOct-1 construct. pCGNOct1 ⌬C was obtained by removing the COOH- (Qiagen, Santa Clarita, CA). Five to 10 ␮g of total RNA/sample were run on terminal fragment (aa 441–743), pCGNOct1 ⌬N was generated by deletion of a 1.2% agarose–formaldehyde gel. RNA was stained in the gel with SYBR aa 1–269, and internal deletion in the POU homeodomain (aa 397–441) Gold (Molecular Probes) and then transferred to a membrane according to the yielded pCGNOct1 ⌬HD. Reporter plasmids pG5E1BCAT, pG5 4xOct standard capillary procedure. The membrane was hybridized with a H2B E1BCAT, and pG5 4xC/EBP E1BCAT have been described previously cDNA probe as described previously (57). ␥-Actin and U2 probes were (52–54). obtained by PCR from a human cDNA library. RNA levels were quantified by Cell Culture, Treatments, and Irradiation. Human HeLa and MCF-7 phosphorimager analysis (Typhoon; Molecular Dynamics) and corrected for cells and Chinese hamster fibroblast cell lines V79 and XR-V15B were loading with the 18S RNA values, with use of ImageQuant software. All cultured in high-glucose DMEM supplemented with 10% fetal bovine serum at experiments included a minimum of three repetitions, and representative phosphoimages are displayed. 37°Cin5%CO2. Human glioma derivatives M059K and -J were maintained in DMEM/F12 supplemented with 10% fetal bovine serum. For irradiation Phosphorylation of Oct-1 in Vivo. V79 and XR-V15B were transfected experiments, cells were plated the night before irradiation at 50–60% conflu- with pCGNOct-1 and incubated for 24 h. Cells were then incubated in ency. Irradiations were performed with a Pantak irradiator at a dose rate of 1.68 phosphate-free medium for 3 h, at which time Zeocin (50 ␮g/ml) and 0.3 32 Gy/min. Irradiation at 2 Gy and above (e.g., 6 and 12 Gy) yielded equivalent mCi/ml PPi were added. After a 30-min incubation, cells were harvested, results. Zeocin treatment was performed by incubating cells in medium con- and whole-cell extracts were prepared. Oct-1 was immunoprecipitated by use taining 50 ␮g/ml Zeocin (Invitrogen) for 1 h. The medium was then replaced of an Oct-1 antibody (C-21; Santa Cruz Biotechnology). by regular medium for further incubation. Wortmannin treatment was done at Immunoprecipitates were resolved by 10% SDS-PAGE and transferred to a 32 5 ␮M Wortmannin (Sigma) for 16 h before irradiation. Caffeine treatment was membrane. After phosphorimager analysis to detect P incorporation, the performed as described previously (55), with use of 3 mM caffeine (Sigma) for membrane was hybridized with an HA antibody (HA11; Covance) to deter- 1 h before irradiation. All experiments were repeated a minimum of three mine Oct-1 levels. Quantification of Oct-1 phosphorylation was done with use times, and quantification was performed as described for individual experi- of ImageQuant software. Phosphorylation levels were calculated as the ratio of 32 ments. the P values and the Oct-1 levels measured by Western analysis. Transfection Assays. Cells were transfected with ExGen 500 (MBI Fer- Phosphorylation of Oct-1 by DNA-PK in Vitro. GST fusion proteins mentas, Burlington, ON, Canada) according to the manufacturer’s conditions. were expressed according to standard protocols (Amersham Biosciences Corp, For Oct-1 protein-expression studies, 100-mm plates were transfected with 0.5 Piscataway, NJ). In vitro DNA-PK assays were performed essentially as ␮g of pCGNOct-1 constructs, except in the experiments described in Fig. 4C, described previously (38). 0.1–0.2 ␮g of protein substrate was incubated for 20 where 0.3 ␮g of pCGN-Oct1 ⌬N were transfected. After transfection, cells min at 30°C in kinase buffer [50 mM HEPES (pH 7.5), 100 mM KCl, 10 mM were incubated for 24 h, and then were trypsinized and divided into three MgCl2, 0.2 mM disodium EGTA] in the presence of 20 ng of DNA, 0.5 units ϩ ␮ ␥ 32 60-mm plates, two of which were subjected to irradiation, with the third used of holo-DNA-PK (Ku DNA-PKcs; Promega), and 5 Ci of [ - P]ATP as the unirradiated control. For CAT assays in MCF-7 cells, 60-mm plates (6000 Ci/mmol; Amersham). Reactions were stopped with SDS loading buffer 32 were transfected with 1 ␮g of CAT reporter construct (pG5E1BCAT or pG5 and loaded on SDS-polyacrylamide gels. P incorporation was visualized by 4xOct E1BCAT) and 0.6 ␮g of a CMV promoter ␤-Gal expression vector. autoradiography or phosphorimager analysis. After transfection, cells were irradiated and returned to the incubator for 18 and 26 h, at which time they were harvested for CAT and ␤-Gal assays, according RESULTS to standard procedures. V79 and XR-V15B were transfected with 0.8 ␮gof CAT reporter construct and 0.1 ␮g of CMV-␤-Gal plasmid and incubated for DNA-PK Is Required for Repression of Histone H2B Expres- 16 h after irradiation before harvesting. Variations in transfection efficiencies sion in Response to IR. To investigate whether the DNA-PK status were normalized with use of the values obtained from the ␤-Gal assay. of cells affected changes in Oct-1-dependent transcription after the Experiments were performed in duplicate or triplicate and repeated a minimum formation of DSBs, we monitored the responses to IR of two Oct-1- of three times. Error bars indicate 1 SD. A t test assuming unequal variance dependent genes, histone H2B and U2, in several cell lines (Fig. 1). was used to assess the statistical difference between control and irradiated samples. The level of significance was set at P ϭ 0.05. Exposure of MCF-7 cells, which retain p53-dependent cell cycle Generation of XR-V15B cells expressing human Ku80 (XR-V15B ϩ Ku80) checkpoints in response to IR (58), to 6 Gy of radiation resulted in a was accomplished by cotransfection of a human Ku80 cDNA expression 90% decrease in H2B mRNA levels over the 24 h after irradiation plasmid (pBJ5 Ku80; Ref. 56) and pRSV-neo at an 8:1 ratio. Forty-eight h after (Fig. 1A), confirming the IR sensitivity of H2B mRNA reported transfection, G418 (400 ␮g/ml; Invitrogen) was added to the medium, and cells previously (50). Over the same period, U2 RNA levels declined were selected for 10 days. Stable transfectants were maintained as a pool in Ͼ60%. The effects observed were not attributable to a general down- medium containing 200 ␮g/ml G418. Extracts from these cells were analyzed regulation of mRNA levels, because ␥-actin mRNA levels in the same by Western blot to confirm the expression of Ku80 and the restoration of the samples remained unchanged throughout the trial. Ku70 protein levels, which were found to be equivalent to the levels observed Experiments with the hamster fibroblast cell line V79 and clonal in the wild-type V79 cells (data not shown). derivative XR-V15B, which lacks Ku80 and in which DNA-PK is Western Blot Analyses. Whole-cell extracts were prepared as described cs inactive (56, 59), demonstrated that H2B and U2 down-regulation was previously (38) in whole-cell extract buffer supplemented with 1 mM sodium vanadate and 20 mM NaF. For Western analysis, extracts were resolved by 8% dependent on an intact DNA-PK kinase complex (Fig. 1B). As in or 10% SDS-PAGE. Gels were transferred to polyvinylidene difluoride mem- MCF-7 cells, H2B and U2 RNA levels declined in the V79 cells over branes (Bio-Rad, Hercules, CA) and hybridized with either of the following the first 6 h after IR, whereas ␥-actin levels were unaffected over this antibodies: Oct-1 (C-21), p53 (FL-393; both from Santa Cruz Biotechnology, time. However, the levels of these RNAs were unaffected by IR in the 7198

fected throughout. Oct-1 was exquisitely sensitive to IR, with a full increase in protein level in response to a dose of only 2 Gy (Fig. 2B). The response was not specific to HeLa cells: we observed a similar rapid and reproducible accumulation of Oct-1 within1hofZeocin treatment of MCF-7 cells (Fig. 2C). Notably, in MCF-7 cells the stabilization of Oct-1 preceded the stabilization of p53. A similar response was observed after treatment with IR (Fig. 3, and data not shown). A second series of experiments indicated that accumulation of Oct-1 in response to the DNA-damage-inducing agents also was

dependent on DNA-PK (Ku and DNA-PKcs; Fig. 3). Addition of the broad phosphatidylinositol-3-kinase inhibitor Wortmannin blocked the response of Oct-1 to IR in HeLa cells (Fig. 3A). In contrast, caffeine, which is a specific inhibitor of ATM and ATR (ATM and Rad3-related protein) but does not affect DNA-PK at the dose used (3 mM; Ref. 55), failed to block the accumulation of Oct-1 in response to IR in MCF-7 cells (Fig. 3B). Caffeine treatment did, however, se- verely blunt the p53 response in MCF-7 cells, as expected for an inhibitor of ATM.

Oct-1 levels also failed to respond to irradiation in DNA-PKcs- and Ku-deficient cells that are devoid of DNA-PK activity (Fig. 3, C and D). Thus the irradiation-induced accumulation of DNA-PK in wild- type M059K glioma cells was not observed in the DNA-PKϪ M059J derivative (Fig. 3C), although steady-state Oct-1 levels were depressed in these cells. Quantification indicated a similar increase in Oct-1 levels at all time points taken between 1 and 8 h after irradiation in wild-type M059K samples (2.5 Ϯ 0.3-fold), whereas no change was detected in the mutant M059J cells (0.97 Ϯ 0.21-fold). Similarly, the up-regulation of Oct-1 in response to irradiation of V79 cells (2.4 Ϯ 0.2-fold) was absent from the Ku80Ϫ XR-V15B cells Fig. 1. Down-regulation of histone H2B mRNA after DNA damage is dependent on Ku and DNA-PKcs. A, MCF-7 cells were irradiated with 6 Gy and harvested after irradiation at the times indicated. Total RNA (5 ␮g) was analyzed by Northern blot with H2B cDNA, U2, and ␥-actin probes as indicated. 18S RNA visualized by SYBR gold staining was used to assess the loading. H2B, U2, and ␥-actin signals were quantified by phosphorimager and the values normalized to the 18S RNA levels. Values below each lane indicate the amount of H2B, ␥-actin, and U2 mRNA relative to the amount obtained for unirradiated cells (Lane C), which was set at 1. B, Northern analysis of H2B, U2, and ␥-actin RNA levels in V79 cells (wild-type; wt), XR-V15B cells (Ku80Ϫ), and a pool of XR-V15B cells stably expressing human Ku80 (V15B ϩ Ku80), prepared as described in the “Materials and Methods,” in response to IR. C, Northern analysis of H2B mRNA levels after irradiation (6 Gy) in M059K (wild-type) and M059J (DNA-PKcs mutant) cells performed as described in the legend for A.

Ku80Ϫ XR-V15B cells. Reconstitution of DNA-PK complex and

DNA-PKcs kinase activity by stable expression of human Ku80 in the XR-V15B cells (56, 59) restored the sensitivity of H2B and U2 mRNA to IR.

A direct requirement for DNA-PKcs for the down-regulation in histone H2B mRNA was also observed in a pair of human glioma cell lines, M059K and M059J (Fig. 1C). M059J cells are a clonal derivative of M059K specifically lacking DNA-PKcs (60). H2B mRNA Ϫ levels were refractory to IR in the DNA-PKcs M059J derivative but repressed in the parental M059K cells, which were wild type for DNA-PK. Accumulation of Oct-1 in Response to DSBs Is Dependent on DNA-PK. To determine whether Oct-1 was directly affected by DSBs, we monitored the response of Oct-1 to IR and the radiomimetic Fig. 2. Analysis of Oct-1 protein levels in response to DSBs. A, HeLa cells were irradiated with 12 Gy and harvested 1, 2, 4, 8, and 24 h postirradiation or after mock compound Zeocin (Fig. 2). In contrast to expectations based on the irradiation (Lane Ϫ) to prepare whole-cell extracts. Ten ␮g of extract were loaded for each depression of H2B and U2 expression observed subsequent to IR, time point on a 10% SDS-polyacrylamide gel, and samples were analyzed by Western blot ␤ Oct-1 levels increased 2.7 Ϯ 0.6-fold in HeLa cells within 1 h after with antibodies to Oct-1, Ku70, and -actin. B, HeLa cells were subjected to 2, 4, and 6 Gy and collected for extract preparation after2hofincubation. Samples were processed irradiation at 12 Gy (Fig. 2A). The elevation in Oct-1 was sustained as in A with hybridization with Oct-1 and actin antibodies. C, MCF-7 cells were incubated for at least 8 h, but subsequently returned to control levels by 24 h. in absence (0h) or presence of 50 ␮g/ml Zeocin for 1 h and collected immediately (Lane 2) or incubated for1h(Lane 3)or2h(Lane 4) in Zeocin-free medium. Samples were Accumulation of Ku70 also increased in response to irradiation, as processed as described in the legend for A, and the membrane was successively hybridized reported previously (61); ␤-actin protein levels, however, were unaf- with antibodies to Oct-1, p53, and actin. 7199

␮ Fig. 3. Response of Oct-1 to IR is dependent on Ku and DNA-PKcs. A, HeLa cells were incubated in the absence (Lanes 1–3) or presence of 5 M Wortmannin (Lanes 4–6) and subjected to 2 or 6 Gy irradiation or left unirradiated (Lane 0). Whole-cell extracts were prepared 2 h after IR treatment, and 10 ␮g of extract were analyzed by Western blot with antibodies to Oct-1 and actin. B, MCF-7 cells were incubated with 3 mM caffeine for1h(Lanes 4–6) or left untreated (Lanes 1–3) before being subjected to 6 Gy irradiation or mock-treated (Lane C). Whole-cell extracts were prepared at the times indicated after irradiation and analyzed by Western blot with successive hybridization of the same membrane with antibodies to Oct-1, p53, and actin. C, M059K cells (DNA-PKcs wild-type) and M059J cells (DNA-PKcs mutant) were irradiated with a dose of 12 Gy and incubated for 1, 4, or 8 h after treatment. Whole-cell lysates were analyzed by Western blot with the antibodies indicated. Oct-1 proteins levels were quantified from samples in a minimum of three experiments, and the amount of Oct-1 in irradiated samples was calculated relative to the mock-irradiated control samples. M059K, 2.54 Ϯ 0.34-fold (n ϭ 11); M059J, 0.97 Ϯ 0.21-fold (n ϭ 10). D, V79 cells (wild-type), XR-V15B cells (Ku80 mutant; KuϪ/Ϫ), and XR-V15B cells stably transfected with human Ku80 cDNA (V15B ϩ Ku80) were harvested after 1 or4hofincubation after treatment with 6 Gy irradiation. Whole-cell extracts were analyzed as described in the legend for C. Quantification yielded the following levels of Oct-1 in irradiated samples relative to control samples. V79, 2.35 Ϯ 0.23-fold (n ϭ 9); V15B, 1.06 Ϯ 0.15-fold (n ϭ 9); V15B ϩ Ku80, 2.3 Ϯ 0.7-fold (n ϭ 8).

(1.1 Ϯ 0.1-fold; Fig. 3D). The IR-induced Oct-1 accumulation was previously shown to lack DNA-PK phosphorylation sites (38), sug- confirmed to be mediated through a Ku-dependent pathway because gested that the effect of DNA-PK on Oct-1 may be direct and reintroduction of wild-type Ku80 in XR-V15B cells restored the dependent on the interaction of Oct-1 with Ku. IR-induced up-regulation of Oct-1 close to levels observed in wild- To determine whether phosphorylation of Oct-1 increased in re- type V79 cells (2.3 Ϯ 0.7-fold). sponse to DNA damage and whether this increase was attributable to 32 An NH2-Terminal Domain of Oct-1 That Is Phosphorylated by DNA-PK, we analyzed PPi incorporation into HA-Oct-1 expressed DNA-PK Mediates Stabilization of Oct-1 in Response to IR. Pre- in wild-type V79 cells and V15B Ku80Ϫ cells treated with Zeocin. A

vious results have shown that Oct-1 mRNA levels are unaffected by 30-min pulse of Pi concomitant with Zeocin treatment showed a exposure to IR (62). To determine whether the change in Oct-1 protein significant increase in the relative level of Oct-1 phosphorylation, at levels we observed in response to IR resulted from changes in Oct-1 1.51 Ϯ 0.09-fold, when phosphorylation was corrected for changes in

stability and to establish a platform for dissection of the effect on the absolute amount of Oct-1 (Fig. 5). In contrast, Pi labeling of Oct-1, we expressed a HA-tagged version of Oct-1 by transient HA-Oct-1 expressed in V15B cells was unaffected by IR transfection (Fig. 4A). In response to IR, HA-tagged Oct-1 accumu- (1.02 Ϯ 0.17-fold), demonstrating that the change in Oct-1 phospho- lated to levels similar to the endogenous octamer factor (2.24 Ϯ 0.2- rylation was dependent upon the presence of holo-DNA-PK. fold), indicating that the up-regulation of Oct-1 in response to IR To determine whether the potential for phosphorylation of Oct-1 by resulted from changes at the protein level rather than the gene level. DNA-PK correlated with the stabilization of Oct-1 in response to IR, To begin to delimit the determinants required for accumulation of we examined the phosphorylation of recombinant Oct-1 constructs by Oct-1 subsequent to IR, we expressed several recombinant HA-Oct-1 purified DNA-PK (Fig. 6). Full-length Oct-1, expressed as a GST constructs in MCF-7 cells (Fig. 4B). The three forms of HA-Oct-1 fusion protein, was efficiently phosphorylated by DNA-PK with ef- expressed containing truncations maintained the Oct-1 nuclear local- ficiency greater than a construct containing only the central region of ization signal and were confirmed by immunofluorescence to be fully Oct-1 including the homeodomain that we had used previously localized to the nucleus of the transfected cells (data not shown). (Fig. 6B).

Interestingly, deletion of the NH2 terminus of Oct-1 to aa 270 resulted Analysis of constructs containing either the NH2- or COOH-termi- in a severalfold elevation in the level of accumulation of the HA- nal part of Oct-1 revealed that phosphorylation was localized to the

tagged protein. In contrast, HA-Oct-1 with a COOH-terminal trunca- NH2-terminal region because the Oct-1 COOH-terminal peptide was tion to aa 440 or containing a partial deletion of the homeodomain that not appreciably phosphorylated by DNA-PK (Fig. 6C). Additionally, abrogates the interaction of Oct-1 with Ku (38) accumulated to levels in this instance, the absence or truncation of the homeodomain from ϳ that were 50% lower than those of full-length HA-Oct-1. the NH2- and COOH-terminal constructs, respectively, prevented the Deletion of the COOH terminus of Oct-1 had no effect on the interaction of these constructs with Ku, leading to a 40-fold reduction response to IR, with HA-Oct-1 ⌬C being stabilized to the same extent in total phosphate incorporation, as we have shown previously for as the endogenous Oct-1 in the same cells (Fig. 4C). In contrast, other homeodomain truncations (38). HA-Oct-1 ⌬N was refractory to IR, suggesting that DNA-PK acted to Addition of the homeodomain to the COOH-terminal region of

stabilize Oct-1 through the NH2 terminus of the protein. Interestingly, Oct-1 did not enhance subsequent phosphorylation. However, attach- deletion of the homeodomain of Oct-1 also abrogated Oct-1 stabili- ment to the NH2-terminal region completely rescued full phosphoryl- zation in response to IR. The effect of this region, which we have ation of this region (data not shown). 7200

motifs lagged significantly 18 h after irradiation, despite the 2-fold increase in Oct-1 levels that were observed in these cells in response to IR (see Fig. 4). Fold activation of EIB expression by the octamer motifs was reduced to half of the level observed in control cells (Fig. 7B). This response was specific to the octamer motif because the C/EBP-dependent activation of transcription from a similar construct containing four copies of a C/EBP response element in lieu of the octamer motifs (52) was not repressed after irradiation (Fig. 7C). At 26 h postirradiation, octamer motif-dependent CAT activity continued to lag the activity observed in mock-irradiated cells. How- ever, the difference between the CAT activity in irradiated and control samples was reduced, consistent with gradual recover in Oct-1 transactivation potential that correlated with the return of Oct-1 to normal levels in the cell.

Fig. 5. In vivo phosphorylation of Oct-1 after DNA damage is dependent on DNA-PK activity. V79 and XR-V15B cells transfected with HA-Oct-1 were incubated for 30 min 32 ϩ Ϫ ␮ with PPi in the presence ( ) or absence ( )of50 g/ml Zeocin. Oct-1 was immunoprecipitated with an Oct-1 antibody, analyzed by phosphorimager to determine 32P-phosphate incorporation (top) and subjected to Western blot analysis with a HA- antibody to determine protein levels (bottom). Note that reduced amounts of Zeocin- treated V79 extracts were used for the immunoprecipitation to account for the increased Fig. 4. Oct-1 response to DNA damage is dependent on the NH2-terminal domain and 32 the homeodomain. A, MCF-7 cells were transfected with pCGNOct-1 expressing a HA Oct-1 levels, as determined by input loading. The relative P-phosphate incorporation in Ϯ ϭ epitope-tagged Oct-1 cDNA. Each transfection was then split equally into three plates, of irradiated samples is indicated below each lane: V79 cells, 1.51 0.09-fold (n 3), in Ϯ ϭ which one was used as control unirradiated (Lane 1) and the other two subjected to a 6-Gy V15B cells: 1.02 0.17-fold (n 3). dose of irradiation and harvested after2h(Lane 2)or4h(Lane 3). Ten ␮g of whole-cell extracts were analyzed by Western blot with antibodies to Oct-1 (which recognize both endogenous and transfected Oct-1), with a HA antibody to assess transfected HA-Oct-1 levels, and with an actin antibody to verify equal loading. B, schematic representation of the wild-type and deletion mutant HA-Oct-1 constructs. The numbering refers to the Oct-1 amino acids included in each construct. The darker gray boxes represent the POU domain (POU-specific and POU homeodomain). The HA epitope is indicated. Shown below is an analysis of the level of expression of each construct in MCF-7 cells after transfection with equal amounts of expression vector for each construct. Ten ␮g of whole-cell extracts were analyzed by Western blot with a HA antibody. C, cells were transfected separately with each construct (⌬C, ⌬HD, and ⌬N), as indicated and processed as described in the legend for A. Ten ␮g of whole-cell extracts were separated by SDS-PAGE, transferred to a membrane, and hybridized with antibodies to Oct-1 to analyze endogenous Oct-1 levels. The membrane was rehybridized with a HA antibody to reveal HA-Oct-1 mutants and with an actin antibody to verify equal loading. Nuclear localization of all constructs was verified by indirect immunofluorescence (data not shown). Quantification yielded the following levels of Oct-1 in irradiated samples relative to control samples. HA-Oct-1 wild-type (Lane C), 2.24 Ϯ 0.2-fold (n ϭ 6); HA-Oct-1⌬C, 1.91 Ϯ 0.27-fold (n ϭ 6); HA-Oct-1⌬HD, 1.01 Ϯ 0.1-fold (n ϭ 6); HA-Oct-1⌬N, 1.08 Ϯ 0.13-fold (n ϭ 6).

Transcriptional Activation by Oct-1 is Selectively Depressed in Response to IR. To investigate how DNA-PK-dependent stabilization of Oct-1 in response to IR correlated with Oct-1-dependent transcription, we monitored Oct-1-dependent reporter gene transcription in transient transfection assays in MCF-7 cells (Fig. 7). To measure transcriptional activation by endogenous Oct-1, we evaluated the effect of four copies of a consensus Oct-1 binding site (H2B octamer motif) on CAT activity derived from an adenovirus E1B Fig. 6. DNA-PK phosphorylates residues present on the NH2-terminal domain of Oct-1. A, schematic representation of the GST-Oct-1 constructs used as substrates in minimal promoter. After the completion of transfection, cells were DNA-PK phosphorylation assays. B, in vitro DNA-PK phosphorylation of purified GST- immediately subjected to 6 Gy of radiation. CAT activity was deter- Oct-1 full-length and GST-Oct-1 POU. After separation of the products by 10% SDS- mined on cells harvested 18 and 26 h thereafter. PAGE, the gel was stained with Coomassie blue (Lanes 1 and 2) and phosphorylation was revealed by Phosphorimager analysis (Lanes 3 and 4). The arrows indicate the positions Transcription from the minimal E1B promoter was unaffected by of the full-size GST-Oct-1 full length and GST-Oct-1 POU proteins. C, Equal amount of irradiation in all instances. ␤-Gal activity from a cotransfected vector GST-Oct-1 C-term (Lanes 1 and 3) or N-term (Lanes 2 and 4) proteins were in vitro phosphorylated by purified DNA-PK. Samples were separated by 10% SDS-PAGE and using a CMV promoter also remained unchanged on IR treatment then stained with Coomassie (left) and autoradiographed to assess 32P incorporation (data not shown). In contrast, CAT activity dependent on the octamer (right). 7201

To assess the role of DNA-PK in the repression of octamer-motif- dependent transcriptional activation, we repeated the transfection experiments in the V79 and XR-V15B cells. In the wild-type V79 cells, the octamer-dependent activation of transcription was again significantly reduced after IR treatment (Fig. 7D). In contrast, reporter gene expression was unaffected by irradiation in the Ku80Ϫ XR-V15B cells. We conclude that DNA-PK acts in response to IR to modulate expression of essential genes such as histone H2B and U2 through the promotion of a decrease in the transcriptional activation potential of Oct-1.

DISCUSSION Our results identify a role for DNA-PK in modulating specific gene transcription in response to DSBs by regulating the levels and activity of Oct-1. These results suggest that the stimulation of DNA-PK activity from DSBs can rapidly and directly affect cellular transcription by targeting transcription factors for phosphorylation. Elevation in the expression of the histone H2B gene is a primary requirement for cellular proliferation because new DNA synthesis must be accompanied by new histone deposition (42, 63). Thus, down-regulation of histone synthesis would seem a natural consequence of conditions in which cell cycle progression is suspended, such as in response to the DNA damage caused by IR. The absence of Ϫ Ϫ down-regulation of H2B in Ku and DNA-PKcs cells provides compelling evidence for a direct role for DNA-PK in regulating this response. Down-regulation of U2 small nuclear RNA, an essential component of the splicing machinery, was found to coincide with the H2B mRNA decrease. U2 expression has been shown to be dependent on Oct-1 binding to the octamer sequence present in the distal sequence element of the U2 promoter (47, 64). Thus, inhibition of Oct-1 transcriptional activity by DNA-PK after IR would be expected to have a direct effect on U2 gene transcription, as observed in our experiments. However, the regulation of U2 transcription is a complex process because full activation of U2 also requires transcription factor Sp1, shown to interact with Oct-1, as well as the assembly of a multisubunit complex recruited to the proximal sequence element (43, 47, 64, 65). Therefore, complete understanding of how U2 is down-regulated after DNA damage will require further investigation.

Several reports have indicated that the absence of Ku or DNA-PKcs does not affect checkpoint activation and has no effect on the inhibition of DNA replication induced by IR (66–69). Indeed, the ATM- dependent S checkpoint activated in response to DNA damage is

stronger in Ku and DNA-PKcs-deficient cells (70–72). However, the Ku- and DNA-PK-deficient cell lines (XR-V15B and M059J), as well as their respective parent cell lines (V79 and M059K), do express a mutated version of p53 (66, 73, 74). Previously it was shown that the Fig. 7. Oct-1-dependent transcription is inhibited after DNA damage. A, MCF-7 cells p53 deficiency does not prevent Oct-1 up-regulation because Oct-1 were transfected with a minimal promoter E1BCAT construct (E1B) or the same construct response to IR was observed in a cell line where the p53 gene is containing four copies of an octamer motif (4xoct E1B). Cells were irradiated (ϩ; 6Gy), or not (Ϫ), and harvested after 18 or 26 h of incubation. Values shown are the means of deleted (62). Furthermore, in the present study we observed a similar triplicate samples from one representative experiment. The error bars indicate the SD of Oct-1 up-regulation after IR in MCF-7 cells, which have normal p53 the mean. B, combined results of three separate experiments performed in duplicate or triplicate. Results are expressed as fold activation of 4xOct E1BCAT activity relative to regulation (58). Lastly, despite the p53 status and the presence of the basal (E1BCAT) promoter activity. Statistical significance: for 18 h, P ϽϽ 0.001; for abnormally low levels of ATM in the M059J cells (75), the Ku- and 26 h, P Ͻ 0.02. C, C/EBP-dependent transcription is not affected by DNA damage. DNA-PK -deficient cells used in this study display normal inhibition MCF-7 cells were transfected with reporter constructs containing either four copies of a cs C/EBP motif adjacent to the E1B minimal promoter or the minimal promoter alone. of DNA synthesis after IR (Refs. 70, 76 and the references therein). Samples were processed as described in the legend for A, and the results of two separate Together, these data indicate that down-regulation of H2B/U2 occurs duplicate experiments are expressed as fold activation as in B. Statistical significance for independently of p53/ATM-dependent checkpoint responses and sug- both 18 and 26 h: P ϾϾ 0.1. D, DNA damage-dependent Oct-1 transcriptional inhibition does not occur in cells lacking DNA-PK activity. V79 (Ku wild-type) and XR-V15B gests that H2B synthesis is uncoupled from DNA replication in the (Ku80Ϫ) cells were transfected with a minimal promoter E1BCAT construct or the same absence of DNA-PK in irradiated cells. construct containing four copies of an octamer motif as in A and harvested 16 h after IR treatment. The combined results of two experiments performed in duplicate are expressed Transcription of the histone H2B gene occurs in a cell-cycle- as fold activation as in B. Statistical significance: for V79, P Ͻ 0.01; for V15B, P ϾϾ 0.1. dependent manner and is largely dependent on Oct-1 (42, 77, 78). Cell 7202

Downloaded from cancerres.aacrjournals.org on September 27, 2021. © 2003 American Association for Cancer Research. Oct-1 TARGETING BY Ku/DNA-PK AFTER DNA DAMAGE cycle control of Oct-1 activity is regulated through phosphorylation of activation domains, including that of the c-myc protein, has been the octamer factor at Ser385 in the homeodomain, which acts by shown to be linked to the rapid turnover of the proteins through the inhibiting Oct-1 DNA binding at the onset of mitosis (79, 80). In 26S proteasome, subsequent to ubiquitin modification within the contrast, DNA-PK modifies Oct-1 in the NH2 terminus, with prelim- transcriptional regulatory domains (84). Although Oct-1 contains a inary results identifying multiple sites of phosphorylation within this glutamine-rich rather than an acidic transcriptional activation domain region. This phosphorylation included sites several amino acids NH2- in its NH2 terminus and its stability in normal cells is severalfold terminal to the POU-specific domain that were phosphorylated in our higher than those of rapidly turned-over proteins such as c-myc (79, previous study using an Oct-1 construct focused on the POU-specific 84), the possibility of an inverse relationship between Oct-1 stability and POU homeodomains of Oct-1 (Ref. 38, and data not shown). and transactivation potential merits further investigation. These results suggest that modification of Oct-1 by DNA-PK is A recurring theme in the study of DNA-PK phosphorylation targets unlikely to affect Oct-1 DNA binding. Furthermore, an increase in is that colocalization of the kinase and substrate through protein– Oct-1 DNA binding activity has been reported after DNA damage, protein and protein–DNA interactions confers a selective advantage to consistent with elevated levels of the protein (50, 62). Thus, the phosphorylation (6, 38, 86–88). In this instance, the phosphorylation modulation of Oct-1 by DNA-PK appears to be distinct from the of Oct-1 and its stabilization in vivo in response to the DNA damage normal cell cycle regulation of the factor. induced by IR correlated directly with the ability of Oct-1 to interact In our experiments, the strong down-regulation of H2B mRNA with the Ku70 subunit of DNA-PK. Furthermore, the interaction of observed somewhat exceeded the 2-fold reduction in transcription Oct-1 with Ku does not appear to lead to the recruitment of Ku to dependent on octamer motifs in a synthetic promoter analyzed by octamer motifs through which Oct-1 acts, but has been demonstrated transient transfection. On the one hand, this suggests the possibility to occur on DNA ends, from which DNA-PK is activated (38), that effects on additional factors that are also involved in regulating supporting the emphasis on the potential for directed phosphorylation H2B expression come together with the effect that we have charac- of Oct-1 in response to DNA damage. Moreover, the interaction with terized on Oct-1 to impart the strong effects observed in vivo. How- Oct-1 containing the homeodomain and the similar interactions dem- ever, it should also be recognized that the nature of the transient onstrated with other homeodomain proteins (38) imply that the acti- transfection assay likely decreases the magnitude of the effect ob- vation of DNA-PK from DNA ends may act broadly to influence the served. Transfection was performed over a period of 5 h, followed by function of homeodomain proteins in the cell in response to IR. irradiation of the cells. Thus it is likely that at least some initial transcriptional response of the promoter occurred over the course of the incubation with transfection agent, before irradiation. The net ACKNOWLEDGMENTS effect of this activity would be to reduce the magnitude of the We are grateful to Winship Herr for providing the pCGNOct-1 plasmid and reduction observed in the level of the very stable CAT protein. to Kyu Lim for the pSPH2B construct. We thank Jeff Kochan and Daniel Similarly, recovery of the cells as seen by the more modest reduction Plouffe for technical assistance. in transcription at the longer time point could also serve to blunt the overall magnitude of the effect observed. It has previously been suggested that Oct-1 induction in response to REFERENCES DNA damage is a post-transcriptional mechanism because Oct-1 1. Abraham, R. T. Cell cycle checkpoint signaling through the ATM and ATR kinases. mRNA levels were not affected by DNA damage and the Oct-1 Genes Dev., 15: 2177–2196, 2001. protein up-regulation was unaltered by the inhibition of RNA synthe- 2. Hartley, K. O., Gell, D., Smith, G. C., Zhang, H., Divecha, N., Connelly, M. A., Admon, A., Lees-Miller, S. P., Anderson, C. W., and Jackson, S. P. DNA-dependent sis (62). We have confirmed here that Oct-1 up-regulation is likely protein kinase catalytic subunit: a relative of phosphatidylinositol 3-kinase and the attributable to a protein modification because ectopically expressed ataxia telangiectasia gene product. Cell, 82: 849–856, 1995. Oct-1 levels were increased similar to the levels of the endogenous 3. Smith, G. C., and Jackson, S. P. The DNA-dependent protein kinase. Genes Dev., 13: Ј 916–934, 1999. protein. The Oct-1 cDNA in our constructs lacks the 5 untranslated 4. Dynan, W. S., and Yoo, S. Interaction of Ku protein and DNA-dependent protein region, which could potentially modulate protein synthesis after IR. In kinase catalytic subunit with nucleic acids. Nucleic Acids Res., 26: 1551–1559, 1998. addition, we can rule out that this up-regulation is attributable to a 5. Giffin, W., Torrance, H., Rodda, D. J., Prefontaine, G. G., Pope, L., and Hache, R. J. Sequence-specific DNA binding by Ku autoantigen and its effects on transcription. potential IR-induced activation of transcription from the CMV pro- Nature (Lond.), 380: 265–268, 1996. moter because we observed a differential stabilization of the mutant 6. Giffin, W., Kwast-Welfeld, J., Rodda, D. J., Prefontaine, G. G., Traykova-Andonova, M., Zhang, Y., Weigel, N. L., Lefebvre, Y. A., and Hache, R. J. Sequence-specific Oct-1 proteins that were expressed from this same promoter. Oct-1 DNA binding and transcription factor phosphorylation by Ku Autoantigen/DNA- protein levels do not fluctuate throughout the cell cycle (79), which dependent protein kinase. Phosphorylation of Ser-527 of the rat glucocorticoid suggests that Oct-1 up-regulation after IR is likely not a consequence receptor. J. Biol. Chem., 272: 5647–5658, 1997. 7. Jackson, S. P. Sensing and repairing DNA double-strand breaks. Carcinogenesis of cell cycle arrest, but rather a direct effect of a DNA damage- (Lond.), 23: 687–696, 2002. induced modification. 8. Karran, P. DNA double strand break repair in mammalian cells. Curr. Opin. Genet. This is further supported by the increased in vivo phosphorylation Dev., 10: 144–150, 2000. 9. Khanna, K. K., and Jackson, S. P. DNA double-strand breaks: signaling, repair and of Oct-1 that occurs after DNA damage. This phosphorylation was the cancer connection. Nat. Genet., 27: 247–254, 2001. rapid (within 30 min) and Ku dependent, suggesting that DNA-PK 10. Ferguson, D. O., and Alt, F. W. DNA double strand break repair and chromosomal translocation: lessons from animal models. Oncogene, 20: 5572–5579, 2001. acts directly to modify Oct-1 stability and transactivation potential. 11. van Gent, D. C., Hoeijmakers, J. H., and Kanaar, R. Chromosomal stability and the It is intriguing that down-regulation of the transactivation potential DNA double-stranded break connection. Nat. Rev. Genet., 2: 196–206, 2001. of Oct-1 occurs concomitant with overall stabilization of the protein. 12. Roth, D. B., and Gellert, M. New guardians of the genome. Nature (Lond.), 404: 823–825, 2000. Initial analysis of the phosphorylation of Oct-1 by DNA-PK within 13. Beamish, H. J., Jessberger, R., Riballo, E., Priestley, A., Blunt, T., Kysela, B., and the NH2 terminus indicates at least three clusters of phosphorylation Jeggo, P. A. The C-terminal conserved domain of DNA-PKcs, missing in the SCID (data not shown), suggesting that the effects on stability and transac- mouse, is required for kinase activity. Nucleic Acids Res., 28: 1506–1513, 2000. 14. Chan, D. W., Chen, B. P., Prithivirajsingh, S., Kurimasa, A., Story, M. D., Qin, J., and tivation may be mediated through separate functional determinants. Chen, D. J. Autophosphorylation of the DNA-dependent protein kinase catalytic However, we also noted with interest recent reports that coupled subunit is required for rejoining of DNA double-strand breaks. Genes Dev., 16: 2333–2338, 2002. ubiquitin-mediated protein turnover with transcriptional activation 15. Kurimasa, A., Kumano, S., Boubnov, N. V., Story, M. D., Tung, C. S., Peterson, (81–85). Transcriptional activation mediated through several acidic S. R., and Chen, D. J. Requirement for the kinase activity of human DNA-dependent 7203

protein kinase catalytic subunit in DNA strand break rejoining. Mol. Cell. Biol., 19: 45. Tanaka, M., Lai, J. S., and Herr, W. Promoter-selective activation domains in Oct-1 3877–3884, 1999. and Oct-2 direct differential activation of an snRNA and mRNA promoter. Cell, 68: 16. Ma, Y., Pannicke, U., Schwarz, K., and Lieber, M. R. Hairpin opening and overhang 755–767, 1992. processing by an Artemis/DNA-dependent protein kinase complex in nonhomologous 46. Yang, J., Muller-Immergluck, M. M., Seipel, K., Janson, L., Westin, G., Schaffner, end joining and V(D)J recombination. Cell, 108: 781–794, 2002. W., and Pettersson, U. Both Oct-1 and Oct-2A contain domains which can activate the 17. Leber, R., Wise, T. W., Mizuta, R., and Meek, K. The XRCC4 gene product is a target ubiquitously expressed U2 snRNA genes. EMBO J., 10: 2291–2296, 1991. for and interacts with the DNA-dependent protein kinase. J. Biol. Chem., 273: 47. Hernandez, N. Small nuclear RNA genes: a model system to study fundamental 1794–1801, 1998. mechanisms of transcription. J. Biol. Chem., 276: 26733–26736, 2001. 18. Kim, S. T., Lim, D. S., Canman, C. E., and Kastan, M. B. Substrate specificities and 48. de Jong, R. N., and van der Vliet, P. C. Mechanism of DNA replication in eukaryotic identification of putative substrates of ATM kinase family members. J. Biol. Chem., cells: cellular host factors stimulating adenovirus DNA replication. Gene, 236: 1–12, 274: 37538–37543, 1999. 1999. 19. O’Neill, T., Dwyer, A. J., Ziv, Y., Chan, D. W., Lees-Miller, S. P., Abraham, R. H., 49. Matheos, D. D., Ruiz, M. T., Price, G. B., and Zannis-Hadjopoulos, M. Oct-1 Lai, J. H., Hill, D., Shiloh, Y., Cantley, L. C., and Rathbun, G. A. Utilization of enhances the in vitro replication of a mammalian autonomously replicating DNA oriented peptide libraries to identify substrate motifs selected by ATM. J. Biol. sequence. J. Cell. Biochem., 68: 309–327, 1998. Chem., 275: 22719–22727, 2000. 50. Meighan-Mantha, R. L., Riegel, A. T., Suy, S., Harris, V., Wang, F. H., Lozano, C., 20. Anderson, C. W. DNA damage and the DNA-activated protein kinase. Trends Whiteside, T. L., and Kasid, U. Ionizing radiation stimulates octamer factor DNA Biochem. Sci., 18: 433–435, 1993. binding activity in human carcinoma cells. Mol. Cell Biochem., 199: 209–215, 1999. 21. Kastan, M. B., and Lim, D. S. The many substrates and functions of ATM. Nat. Rev. 51. Tanaka, M., and Herr, W. Differential transcriptional activation by Oct-1 and Oct-2: Mol. Cell. Biol., 1: 179–186, 2000. interdependent activation domains induce Oct-2 phosphorylation. Cell, 60: 375–386, 22. Nyberg, K. A., Michelson, R. J., Putnam, C. W., and Weinert, T. A. Toward 1990. maintaining the genome: DNA damage and replication checkpoints. Annu. Rev. 52. Boruk, M., Savory, J. G., and Hache, R. J. AF-2-dependent potentiation of CCAAT Genet., 36: 617–656, 2002. enhancer binding protein beta-mediated transcriptional activation by glucocorticoid 23. Woo, R. A., Jack, M. T., Xu, Y., Burma, S., Chen, D. J., and Lee, P. W. DNA receptor. Mol. Endocrinol., 12: 1749–1763, 1998. damage-induced apoptosis requires the DNA-dependent protein kinase, and is medi- 53. Lillie, J. W., and Green, M. R. Transcription activation by the adenovirus E1a protein. ated by the latent population of p53. EMBO J., 21: 3000–3008, 2002. Nature (Lond.), 338: 39–44, 1989. 24. Wang, S., Guo, M., Ouyang, H., Li, X., Cordon-Cardo, C., Kurimasa, A., Chen, D. J., 54. Pre´fontaine, G. G., Lemieux, M. E., Giffin, W., Schild-Poulter, C., Pope, L., LaCasse, Fuks, Z., Ling, C. C., and Li, G. C. The catalytic subunit of DNA-dependent protein E., Walker, P., and Hache´, R. J. Recruitment of octamer transcription factors to DNA kinase selectively regulates p53-dependent apoptosis but not cell-cycle arrest. Proc. by glucocorticoid receptor. Mol. Cell Biol., 18: 3416–3430, 1998. Natl. Acad. Sci. USA, 97: 1584–1588, 2000. 55. Sarkaria, J. N., Busby, E. C., Tibbetts, R. S., Roos, P., Taya, Y., Karnitz, L. M., and 25. Basu, S., Rosenzweig, K. R., Youmell, M., and Price, B. D. The DNA-dependent Abraham, R. T. Inhibition of ATM and ATR kinase activities by the radiosensitizing protein kinase participates in the activation of NF␬B following DNA damage. agent, caffeine. Cancer Res., 59: 4375–4382, 1999. Biochem. Biophys. Res. Commun., 247: 79–83, 1998. 56. Smider, V., Rathmell, W. K., Lieber, M. R., and Chu, G. Restoration of X-ray 26. Burma, S., Chen, B. P., Murphy, M., Kurimasa, A., and Chen, D. J. ATM phospho- resistance and V(D)J recombination in mutant cells by Ku cDNA. Science (Wash. rylates histone H2AX in response to DNA double-strand breaks. J. Biol. Chem., 276: DC), 266: 288–291, 1994. 42462–42467, 2001. 57. Lee, M. S., Son, M. Y., Park, J. I., Park, C., Lee, Y. C., Son, C. B., Kim, Y. S., Paik, 27. Wang, H., Guan, J., Perrault, A. R., Wang, Y., and Iliakis, G. Replication protein A2 S. G., Yoon, W. H., Park, S. K., Hwang, B. D., and Lim, K. Modification of octamer phosphorylation after DNA damage by the coordinated action of ataxia telangiectasia- binding transcriptional factor is related to H2B histone gene repression during mutated and DNA-dependent protein kinase. Cancer Res., 61: 8554–8563, 2001. dimethyl sulfoxide-dependent differentiation of HL-60 cells. Cancer Lett., 172: 28. Soubeyrand, S., Pope, L., Pakuts, B., and Hache´, R. J. Threonines 2638/2647 in 165–170, 2001. DNA-PK are essential for cellular resistance to ionizing radiation. Cancer Res., 63: 58. Fan, S., Smith, M. L., Rivet, D. J. 2nd, Duba, D., Zhan, Q., Kohn, K. W., Fornace, 1198–1201, 2003. A. J., Jr., and O’Connor, P. M. Disruption of p53 function sensitizes breast cancer 29. Datta, R., Weichselbaum, R., and Kufe, D. W. Ionizing radiation down-regulates MCF-7 cells to cisplatin and pentoxifylline. Cancer Res., 55: 1649–1654, 1995. histone H1 gene expression by transcriptional and post-transcriptional mechanisms. 59. Errami, A., Smider, V., Rathmell, W. K., He, D. M., Hendrickson, E. A., Zdzienicka, Radiat Res., 133: 176–181, 1993. M. Z., and Chu, G. Ku86 defines the genetic defect and restores X-ray resistance and 30. Hain, J., Crompton, N. E., Burkart, W., and Jaussi, R. Caffeine release of radiation V(D)J recombination to complementation group 5 hamster cell mutants. Mol. Cell induced S and G2 phase arrest in V79 hamster cells: increase of histone messenger Biol., 16: 1519–1526, 1996. RNA levels and p34cdc2 activation. Cancer Res., 53: 1507–1510, 1993. 60. Lees-Miller, S. P., Godbout, R., Chan, D. W., Weinfeld, M., Day, R. S. R., Barron, 31. Woloschak, G. E., Felcher, P., and Chang-Liu, C. M. Combined effects of ionizing G. M., and Allalunis-Turner, J. Absence of p350 subunit of DNA-activated protein radiation and cycloheximide on gene expression. Mol. Carcinog., 13: 44–49, 1995. kinase from a radiosensitive human cell line. Science (Wash. DC), 267: 1183–1185, 32. Tuteja, R., and Tuteja, N. Ku autoantigen: a multifunctional DNA-binding protein. 1995. Crit. Rev. Biochem. Mol. Biol., 35: 1–33, 2000. 61. Brown, K. D., Lataxes, T. A., Shangary, S., Mannino, J. L., Giardina, J. F., Chen, J., 33. Camara-Clayette, V., Thomas, D., Rahuel, C., Barbey, R., Cartron, J. P., and and Baskaran, R. Ionizing radiation exposure results in up-regulation of Ku70 via a Bertrand, O. The repressor which binds the Ϫ75 GATA motif of the GPB promoter p53/ataxia-telangiectasia-mutated protein-dependent mechanism. J. Biol. Chem., 275: contains Ku70 as the DNA binding subunit. Nucleic Acids Res., 27: 1656–1663, 6651–6656, 2000. 1999. 62. Zhao, H., Jin, S., Fan, F., Fan, W., Tong, T., and Zhan, Q. Activation of the 34. Ghoshal, K., Li, Z., and Jacob, S. T. Overexpression of the large subunit of the protein transcription factor Oct-1 in response to DNA damage. Cancer Res., 60: 6276–6280, Ku suppresses metallothionein-I induction by heavy metals. Proc. Natl. Acad. Sci. 2000. USA, 95: 10390–10395, 1998. 63. Osley, M. A. The regulation of histone synthesis in the cell cycle. Annu. Rev. 35. Nishishita, T., Okazaki, T., Ishikawa, T., Igarashi, T., Hata, K., Ogata, E., and Fujita, Biochem., 60: 827–861, 1991. T. A negative vitamin D response DNA element in the human parathyroid hormone- 64. Janson, L., Weller, P., and Pettersson, U. Nuclear factor I can functionally replace related peptide gene binds to vitamin D receptor along with Ku antigen to mediate transcription factor Sp1 in a U2 small nuclear RNA gene enhancer. J. Mol. Biol., 205: negative gene regulation by vitamin D. J. Biol. Chem., 273: 10901–10907, 1998. 387–396, 1989. 36. Willis, D. M., Loewy, A. P., Charlton-Kachigian, N., Shao, J. S., Ornitz, D. M., and 65. Strom, A. C., Forsberg, M., Lillhager, P., and Westin, G. The transcription factors Sp1 Towler, D. A. Regulation of osteocalcin gene expression by a novel Ku antigen and Oct-1 interact physically to regulate human U2 snRNA gene expression. Nucleic transcription factor complex. J. Biol. Chem., 277: 37280–37291, 2002. Acids Res., 24: 1981–1986, 1996. 37. Jeanson, L., and Mouscadet, J. F. Ku represses the HIV-1 transcription: identification 66. Allalunis-Turner, J., Barron, G. M., and Day, R. S. 3rd. Intact G2-phase checkpoint of a putative Ku binding site homologous to the mouse mammary tumor virus NRE1 in cells of a human cell line lacking DNA-dependent protein kinase activity. Radiat. sequence in the HIV-1 long terminal repeat. J. Biol. Chem., 277: 4918–2494, 2002. Res., 147: 284–287, 1997. 38. Schild-Poulter, C., Pope, L., Giffin, W., Kochan, J. C., Ngsee, J. K., Traykova- 67. Gu, Y., Seidl, K. J., Rathbun, G. A., Zhu, C., Manis, J. P., van der Stoep, N., Andonova, M., and Hache´, R. J. G. The binding of Ku antigen to homeodomain Davidson, L., Cheng, H. L., Sekiguchi, J. M., Frank, K., Stanhope-Baker, P., proteins promotes their phosphorylation by DNA-dependent protein kinase. J. Biol. Schlissel, M. S., Roth, D. B., and Alt, F. W. Growth retardation and leaky SCID Chem., 276: 16848–16856, 2001. phenotype of Ku70-deficient mice. Immunity, 7: 653–665, 1997. 39. Phillips, K., and Luisi, B. The virtuoso of versatility: POU proteins that flex to fit. 68. Nussenzweig, A., Chen, C., da Costa Soares, V., Sanchez, M., Sokol, K., J. Mol. Biol., 302: 1023–1039, 2000. Nussenzweig, M. C., and Li, G. C. Requirement for Ku80 in growth and immuno- 40. Ryan, A. K., and Rosenfeld, M. G. POU domain family values: flexibility, partner- globulin V(D)J recombination. Nature (Lond.), 382: 551–555, 1996. ships, and developmental codes. Genes Dev., 11: 1207–1225, 1997. 69. Lim, D. S., Vogel, H., Willerford, D. M., Sands, A. T., Platt, K. A., and Hasty, P. 41. Verrijzer, C. P., and Van der Vliet, P. C. POU domain transcription factors. Biochim. Analysis of ku80-mutant mice and cells with deficient levels of p53. Mol. Cell Biol., Biophys. Acta, 1173: 1–21, 1993. 20: 3772–3780, 2000. 42. Heintz, N. The regulation of histone gene expression during the cell cycle. Biochim. 70. Guan, J., DiBiase, S., and Iliakis, G. The catalytic subunit DNA-dependent protein Biophys. Acta, 1088: 327–339, 1991. kinase (DNA-PKcs) facilitates recovery from radiation-induced inhibition of DNA 43. Janson, L., Bark, C., and Pettersson, U. Identification of proteins interacting with the replication. Nucleic Acids Res., 28: 1183–1192, 2000. enhancer of human U2 small nuclear RNA genes. Nucleic Acids Res., 15: 4997– 71. Jeggo, P. A. X-ray sensitive mutants of Chinese hamster ovary cell line: radio- 5016, 1987. sensitivity of DNA synthesis. Mutat. Res., 145: 171–176, 1985. 44. Tanaka, M., Grossniklaus, U., Herr, W., and Hernandez, N. Activation of the U2 72. Zhou, X. Y., Wang, X., Wang, H., Chen, D. J., Li, G. C., Iliakis, G., and Wang, Y. snRNA promoter by the octamer motif defines a new class of RNA polymerase II Ku affects the ATM-dependent S phase checkpoint following ionizing radiation. enhancer elements. Genes Dev., 2: 1764–1778, 1988. Oncogene, 21: 6377–6381, 2002. 7204

73. Anderson, C. W., and Allalunis-Turner, M. J. Human TP53 from the malignant 81. Chi, Y., Huddleston, M. J., Zhang, X., Young, R. A., Annan, R. S., Carr, S. A., and glioma-derived cell lines M059J and M059K has a cancer-associated mutation in Deshaies, R. J. Negative regulation of Gcn4 and Msn2 transcription factors by Srb10 exon 8. Radiat. Res., 154: 473–476, 2000. cyclin-dependent kinase. Genes Dev., 15: 1078–1092, 2001. 74. Chaung, W., Mi, L. J., and Boorstein, R. J. The p53 status of Chinese hamster V79 82. Fryer, C. J., Lamar, E., Turbachova, I., Kintner, C., and Jones, K. A. Mastermind cells frequently used for studies on DNA damage and DNA repair. Nucleic Acids mediates chromatin-specific transcription and turnover of the Notch enhancer com- Res., 25: 992–994, 1997. plex. Genes Dev., 16: 1397–1411, 2002. 75. Chan, D. W., Gately, D. P., Urban, S., Galloway, A. M., Lees-Miller, S. P., Yen, T., 83. Molinari, E., Gilman, M., and Natesan, S. Proteasome-mediated degradation of and Allalunis-Turner, J. Lack of correlation between ATM protein expression and transcriptional activators correlates with activation domain potency in vivo. EMBO J., tumour cell radiosensitivity. Int. J. Radiat. Biol., 74: 217–224, 1998. 18: 6439–6447, 1999. 76. Verhaegh, G. W., Jongmans, W., Morolli, B., Jaspers, N. G., van der Schans, G. P., 84. Salghetti, S. E., Muratani, M., Wijnen, H., Futcher, B., and Tansey, W. P. Functional Lohman, P. H., and Zdzienicka, M. Z. A novel type of X-ray-sensitive Chinese overlap of sequences that activate transcription and signal ubiquitin-mediated prote- hamster cell mutant with radioresistant DNA synthesis and hampered DNA double- olysis. Proc. Natl. Acad. Sci. USA, 97: 3118–3123, 2000. strand break repair. Mutat. Res., 337: 119–129, 1995. 85. Salghetti, S. E., Caudy, A. A., Chenoweth, J. G., and Tansey, W. P. Regulation of 77. Fletcher, C., Heintz, N., and Roeder, R. G. Purification and characterization of OTF-1, a transcription factor regulating cell cycle expression of a human histone H2b transcriptional activation domain function by ubiquitin. Science (Wash. DC), 293: gene. Cell, 51: 773–781, 1987. 1651–1653, 2001. 78. LaBella, F., Sive, H. L., Roeder, R. G., and Heintz, N. Cell-cycle regulation of a 86. Gottlieb, T. M., and Jackson, S. P. The DNA-dependent protein kinase—requirement human histone H2b gene is mediated by the H2b subtype-specific consensus element. for DNA ends and association with ku antigen. Cell, 72: 131–142, 1993. Genes Dev., 2: 32–39, 1988. 87. Soubeyrand, S., Torrance, H., Giffin, W., Gong, W., Schild-Poulter, C., and Hache, 79. Roberts, S. B., Segil, N., and Heintz, N. Differential phosphorylation of the transcription R. J. Activation and autoregulation of DNA-PK from structured single-stranded DNA factor Oct1 during the cell cycle. Science (Wash. DC), 253: 1022–1026, 1991. and coding end hairpins. Proc. Natl. Acad. Sci. USA, 98: 9605–9610, 2001. 80. Segil, N., Roberts, S. B., and Heintz, N. Mitotic phosphorylation of the Oct-1 88. Yannone, S. M., Roy, S., Chan, D. W., Murphy, M. B., Huang, S., Campisi, J., and homeodomain and regulation of Oct-1 DNA binding activity. Science (Wash. DC), Chen, D. J. Werner syndrome protein is regulated and phosphorylated by DNA- 254: 1814–1816, 1991. dependent protein kinase. J. Biol. Chem., 276: 38242–38248, 2001.

7205

Downloaded from cancerres.aacrjournals.org on September 27, 2021. © 2003 American Association for Cancer Research. Down-Regulation of Histone H2B by DNA-Dependent Protein Kinase in Response to DNA Damage through Modulation of Octamer Transcription Factor 1

Caroline Schild-Poulter, Amy Shih, Nicholas C. Yarymowich, et al.

Cancer Res 2003;63:7197-7205.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/63/21/7197

Cited articles This article cites 88 articles, 41 of which you can access for free at: http://cancerres.aacrjournals.org/content/63/21/7197.full#ref-list-1

Citing articles This article has been cited by 11 HighWire-hosted articles. Access the articles at: http://cancerres.aacrjournals.org/content/63/21/7197.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/63/21/7197. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.