DNA-PK Phosphorylation Sites on Oct-1 Promote Cell Survival Following DNA Damage
Total Page:16
File Type:pdf, Size:1020Kb
Oncogene (2007) 26, 3980–3988 & 2007 Nature Publishing Group All rights reserved 0950-9232/07 $30.00 www.nature.com/onc ORIGINAL ARTICLE DNA-PK phosphorylation sites on Oct-1 promote cell survival following DNA damage C Schild-Poulter1,4, A Shih1, D Tantin3,5, NC Yarymowich1, S Soubeyrand1, PA Sharp3 and RJG Hache´ 1,2 1Department of Medicine, The Ottawa Health Research Institute, University of Ottawa, Ottawa, Ontario, Canada; 2Department of Biochemistry, Microbiology and Immunology, The Ottawa Health Research Institute, University of Ottawa, Ottawa, Ontario, Canada and 3Center for Cancer Research and Department of Biology, Massachusetts Institute of Technology, Cambridge, MA, USA Octamer transcription factor-1 (Oct-1) has recently been Introduction shown to function as a stress sensor that promotes cell survival subsequent to DNA damage. Here, we show that Cells that are subjected to double-stranded breaks the survival signal imparted by Oct-1 following exposure (DSBs) in their DNA activate cell cycle checkpoints to ionizing radiation (IR) is dependent upon DNA- and alter specific gene transcription patterns to allow the dependent protein kinase (DNA-PK)-dependent phospho- DNA repair machinery to correct the damage and rylation ofa cluster of13 specific ser/thr residues within preserve the integrity of the genome (Valerie and Povirk, the N-terminal transcriptional regulatory domain of 2003). The predominant DSBs repair pathway in higher Oct-1. Although IR treatment did not affect the recruit- eukaryotic cells is the non-homologous DNA end- ment ofOct-1 to the histone H2B promoter, the joining pathway (NHEJ) (Lees-Miller and Meek, 2003; recruitment ofRNA polymerase II, TATA-binding Lieber et al., 2003). Defects in DNA-break repair are a protein and histone H4 acetylation were strongly reduced, major cause of genomic instability that leads to the consistent with a decrease in Oct-1 transcriptional development of cancer. regulatory potential following IR exposure. Ser/Thr-Ala A key NHEJ factor is the DNA-dependent protein substitution of13 sites present in Oct-1 transcriptional kinase (DNA-PK). DNA-PK is comprised of a large regulatory domain eliminated Oct-1 phosphorylation catalytic subunit, DNA-PKcs, and the Ku antigen subsequent to IR exposure. Further, these substitutions heterodimer (Ku70/Ku80) (Smith and Jackson, 1999). prevented Oct-1 from rescuing the survival of IR-treated DNA-PK is essential in maintaining genomic integrity, Oct-12/2 murine embryonic fibroblasts, providing a direct as mice lacking either DNA-PKcs or Ku suffer from link between DNA-PK-dependent phosphorylation and the profound radiosensitivity, immunodeficiency and age contribution ofOct-1 to cell survival. These results prematurely (Ferguson and Alt, 2001; Lieber et al., implicate Oct-1 as a primary effector in a DNA-PK- 2003). At the cellular level, DNA-PK deficiency results dependent cell survival pathway that is activated by in defective repair of DSBs and defects in telomere double-stranded DNA breaks. maintenance that lead to telomeric fusions and translo- Oncogene (2007) 26, 3980–3988; doi:10.1038/sj.onc.1210165; cations (Lieber et al., 2003; Bailey and Goodwin, 2004; published online 8 January 2007 Collis et al., 2005). DNA-PK exhibits Ser/Thr kinase activity and dis- Keywords: Ku; DNA-dependent protein kinase; Oct-1; plays a strong preference for S/T-Q motifs, although S/ DNA damage; phosphorylation T’s flanked by residues other than Q are also recognized on occasion (Chan et al., 1999; Kim et al., 1999). Suggested substrates for DNA-PK include several factors of the NHEJ pathway (Meek et al., 2004; Collis et al., 2005). DNA-PK is also known to autophos- Correspondence: Dr RJG Hache´ , Hormones, Growth and Develop- phorylate (Chan et al., 2002; Douglas et al., 2002; ment, The Ottawa Health Research Institute, 725 Parkdale Avenue, Soubeyrand et al., 2003). A role for DNA-PK in Ottawa, Ontario, Canada K1Y 4E9. Dr C Schild-Poulter, Robarts signaling to apoptosis machinery has been documented Research Institute, P.O. Box 5015, 100, Perth Drive, London, Ontario, (Bharti et al., 1998; Wang et al., 2000; Woo et al., 2002; Canada N6A 5K8. Sawada et al., 2003). However, the potential for DNA- E-mail: [email protected] or [email protected] 4Current address: Cell Biology Research Group, Robarts Research PK to act as a signaling molecule in regulating other Institute, PO Box 5015, 100, Perth Drive, London, Ontario, Canada cellular responses to DSBs remains to be elucidated. N6A 5K8. Octamer transcription factor-1 (Oct-1) has recently 5Current address: Department of Pathology, University of Utah emerged as an important contributor to cell survival in School of Medicine, EEJB-565, Room 5700B, 15 No. Medical Drive East, Salt Lake City, UT 84112-5650, USA. response to various types of DNA damage (Tantin et al., Received 10 April 2006; revised 29 August 2006; accepted 1 September 2005). Somewhat paradoxically, we have determined 2006; published online 8 January 2007 that Oct-1 protein levels increase subsequent to ionizing Modulation of cell survival through Oct-1 phosphorylation C Schild-Poulter et al 3981 radiation (IR) exposure, whereas at the same time H2B a IR and U2 expression is down-regulated (Zhao et al., C 1.5h 5h 2000a; Jin et al., 2001; Schild-Poulter et al., 2003). Oct-1 is a ubiquitous transcription factor of the Input POU (Pit/Oct-1/2-Unc) subfamily of homeodomain Oct-1 proteins (Herr and Cleary, 1995; Ryan and Rosenfeld, 1997) and is a key regulator of housekeeping genes TBP such as H2B and U2 (Sive and Roeder, 1986; Fletcher Ac H4 et al., 1987). Oct-1 also participates in the transcrip- tional regulation of many signal-regulated and tissue- Pol II specific genes and in the control of viral replication Gal4 (Herr and Cleary, 1995; Ryan and Rosenfeld, 1997; de Jong and van der Vliet, 1999; Phillips and Luisi, b 2000). Oct-1 is essential for embryonic development H2B as Oct-1-deficient embryos die during gestation (Wang 18S et al., 2004). Recently, we determined that the stabilization of c Total Oct-1 in response to IRcorrelated with an overall molar Ac H4 increase in Oct-1 phosphorylation and physical interac- tion with Ku (Schild-Poulter et al., 2003). In vitro Figure 1 Oct-1’s presence on the H2B promoter is unaffected by experiments showed that the association of Oct-1 with DNA damage. (a) ChIP analysis of the H2B promoter in MCF-7 cells untreated (C) and harvested 1.5 h or 5 h after 6 Gy irradiation. Ku facilitated the phosphorylation of Oct-1 by DNA- Cross-linked protein–DNA complexes were immunoprecipitated PK (Schild-Poulter et al., 2001, 2003). with antibodies to the proteins indicated and the purified DNA was Here, we have identified 15 specific phosphorylation analysed by polymerase chain reaction with primers amplifying the sites for DNA-PK on Oct-1, 13 of which occur within H2B promoter. Input corresponds to 0.5–1% of equivalent ChIP samples. (b) Northern blot analysis of histone H2B mRNA after IR the N-terminal glutamine-rich Oct-1 transcriptional treatment. MCF-7 cells were irradiated and harvested as above. regulatory domain. Substitution of these 13 residues The RNA was analysed by Northern blot with a H2B cDNA probe with alanine incrementally prevented Oct-1 stabilization and 18S RNA was visualized by SYBR gold staining. (c) Total in response to IRand abrogated the IR-induced Oct-1 acetylated histone H4 levels were analysed by Western blot using phosphorylation previously shown to be dependent on extracts from cells treated as described in panel a. DNA-PK. Notably, the alanine-substituted Oct-1 failed to rescue the radiation sensitivity of Oct-1À/À mouse embryonic fibroblasts (MEFs) seen upon re-expression not a general effect of DNA damage as total acetylated of wild-type (WT) Oct-1. These results provide the first histone H4 levels remain unchanged (Figure 1c). evidence of a role for DNA-PK as a signaling molecule By contrast, the down-regulation of H2B transcrip- in promoting cell survival in response to DSBs through tion in response to IRwas accomplished without modulation of the activity of the transcription factor affecting the promoter occupancy of Oct-1 (Figure 1a). Oct-1. Thus, the IR-dependent modification of Oct-1 that leads to its stabilization did not appear to affect its targeting to chromatin. This result was consistent with previous observations that Oct-1 DNA-binding to an octamer Results motif was unaffected in extracts prepared from cells subjected to IRtreatment (Meighan-Mantha et al., Oct-1 binding to the H2B promoter is unaffected by 1999; Zhao et al., 2000a). This suggested that the salient DNA damage phosphorylation sites for the IRresponse were outside Oct-1 is readily visualized on the H2B promoter in of the POU DNA-binding domain of Oct-1 that proliferating cells (Zhao et al., 2000b). In response to IR interacts with Ku. treatment, H2B transcription decreases (Meighan- Mantha et al., 1999; Schild-Poulter et al., 2003; Su et al., 2004), suggesting that phosphorylation of Oct-1 in DNA-PKphosphorylates Oct-1 at multiple sites in vitro response to IRcould influence the recruitment of The Oct-1 POU domain is flanked by a N-terminal Oct-1 to the histone H2B promoter. To assess how IR region exhibiting a strong Q-rich character and a exposure affected Oct-1-dependent transcription in vivo, C-terminus containing a S/T-rich region (Figure 2a). we performed a chromatin immunoprecipitation (ChIP) To investigate the hypothesis that the DNA-PK- analysis of the H2B promoter (Figure 1a). The decrease dependent changes in Oct-1 subsequent to treatment in histone H2B transcription upon IRexposure with DNA-damaging agents resulted from direct phos- (Figure 1b) correlated with a rapid decrease in the phorylation, we initiated a characterization of the Oct-1 presence of RNA polymerase II and TATA box-binding residues phosphorylated by purified preparations of protein (Figure 1a).