ã
Oncogene (2001) 20, 5964 ± 5971 2001 Nature Publishing Group All rights reserved 0950 ± 9232/01 $15.00 www.nature.com/onc
Induction of ser15 and lys382 modi®cations of p53 by blockage of transcription elongation Mats Ljungman*,1,2, Heather M O'Hagan1,2 and Michelle T Paulsen1 1
Department of Radiation Oncology, Division of Cancer Biology, University of Michigan Comprehensive Cancer Center, Ann Arbor, Michigan, MI 48109-0936, USA; 2Program in Cellular and Molecular Biology, University of Michigan Medical School, Ann Arbor, Michigan, MI 48109-0936, USA
Blockage of transcription has been shown to induce the tumor suppressor p53 in human cells. We here show that RNA synthesis inhibitors blocking the phosphorylation of the carboxyl terminal domain (CTD) of RNA polymerase II, such as DRB and H7, induced rapid nuclear accumulation of p53 proteins that were not phosphorylated at ser15 or acetylated at lys382. In contrast, agents that inhibit the elongation phase of transcription, such as UV light, camptothecin or actinomycin D, induced the accumulation of nuclear p53 proteins that were modi®ed at both of these sites. Furthermore, using a panel of DNA repair-de®cient cells we show that persistent DNA lesions in the transcribed strand of active genes are responsible for the induction of the ser15 and lys382 modi®cations following UV-irradiation. We conclude that inhibition of transcription is sucient for the accumulation of p53 in the nucleus regardless of whether the ser15 site of p53 is phosphorylated or not. Importantly, blockage of the elongation phase of transcription triggers a distinct signaling pathway leading to p53 modi®cations on ser15 and lys382. We propose that the elongating RNA polymerase complex may act as a sensor of DNA damage and as an integrator of cellular stress signals. Oncogene (2001) 20, 5964 ± 5971. Keywords: RNA polymerase II; inhibitors; DNA damage; UV light; Xeroderma pigmentosum; Cockayne's syndrome Introduction Transcription is a fundamental process for all living organisms. Attenuation of transcription by either the induction of DNA damage or by inhibition of RNA polymerase activity seriously aects many functions of a cell. It has been shown that inhibition of transcription triggers the induction of p53 (Ljungman and *Correspondence: M Ljungman, Department of Radiation Oncology, Division of Cancer Biology, University of Michigan Comprehensive Cancer Center, 4306 CCGC, 150 E. Medical Center Drive, Ann Arbor, MI 48109-0936, USA; E-mail:
[email protected] Received 16 March 2001; revised 7 June 2001; accepted 14 June 2001
Zhang, 1996; Ljungman et al., 1999; McKay et al., 1998; Yamaizumi and Sugano, 1994). However, the mechanism by which p53 proteins accumulate following inhibition of transcription is not clear (Ljungman, 2000). It has also been shown that prolonged inhibition of mRNA synthesis may result in apoptosis (Koumenis and Giaccia, 1997; Ljungman and Zhang, 1996; Ljungman et al., 1999; McKay and Ljungman, 1999; McKay et al., 1998, 2000; Poele et al., 1999). Thus, it is of paramount importance for the cell to rapidly respond to transcriptional stress and to restore transcriptional activity. It is thought that phosphorylation of certain Nterminal residues of p53 may be important for nuclear accumulation of p53 following cellular stresses. These phosphorylations are thought to attenuate the interaction between p53 and its negative regulator the MDM2 protein. MDM2 is thought to direct the nuclear export and subsequent degradation of the p53 protein (Boyd et al., 2000; Chehab et al., 1999; Geyer et al., 2000; Shieh et al., 1997; Siliciano et al., 1997; Unger et al., 1999a). Although phosphorylation of the ser15 site of p53 is not sucient to directly interfere with MDM2 binding to stabilize the p53 protein (Dumaz and Meek, 1999), phosphorylation of this site by the ATM, ATR or p38 protein kinases (Banin et al., 1998; Canman et al., 1998; She et al., 2000; Tibbetts et al., 1999) have been shown to trigger further modi®cation cascades of p53 (Ito et al., 2001; Sakaguchi et al., 1998; 2000). Furthermore, phosphorylation of ser15 has been shown to induce the binding of p300/CBP to p53 resulting in increased transactivation ability of the p53 protein (Dumaz and Meek, 1999; Lambert et al., 1998). The acetylation of the lys382 site of p53 is thought to increase its sequence-speci®c DNA-binding activity (Gu and Roeder, 1997) as well as its protein stability (Ito et al., 2001; Yuan et al., 1999). Taken together, the modi®cations of ser15 and lys382 are thought to be important for stimulating p53-mediated transactivation, cell cycle arrest and apoptosis (Dumaz and Meek, 1999; Fiscella et al., 1995; Unger et al., 1999b). In addition to the role of protein modi®cations in inhibiting the negative regulation of MDM2 following induction of cellular stress, certain stresses may aect the expression of the MDM2 gene by inhibiting general transcription. Following exposure to high doses of UV light, for example, the expression of MDM2 is severely
Transcription blockage and p53 signaling M Ljungman et al
attenuated (Lu et al., 1996; Perry et al., 1993; Reinke and Lozano, 1997; Wang et al., 1999). This is most likely due to the formation of transcription-blocking DNA lesions in the transcribed strand of the Mdm2 gene. Furthermore, agents that are known to inhibit transcription such as actinomycin D, H7 and DRB induce the accumulation of p53 concomitant with down-regulation of MDM2 protein expression (An et al., 1998; Ashcroft et al., 2000; Blattner et al., 1999a). Thus, it is possible that blockage of RNA polymerase II may trigger p53 accumulation by an indirect mechanism involving the attenuation of MDM2 expression. In this study, we investigated whether the p53 proteins accumulating following inhibition of transcription are modi®ed at the ser15 and/or lys382 sites and if they localize to the cell nucleus. Furthermore, using cells with various defects in nucleotide excision repair we explored whether transcription blockage by persistent lesions in the transcribed strand of active genes rather than membrane-mediated events is responsible for triggering the modi®cations at the ser15 and lys382 sites of p53 following UV-irradiation. Our results suggest that blockage of transcription elongation, but not inhibition of promoter escape, results in the induction of ser15 phosphorylation and lys382 acetylation of p53. Thus, accumulation of p53 following inhibition of transcription elongation may not solely be due to attenuated MDM2 expression but may also result from a stress pathway triggered by blockage of the transcription elongation complex.
on DNA in a cleavable complex (Chen and Liu, 1994) and actinomycin D intercalates DNA (Sobell, 1985). The resulting DNA lesions induced by these three agents inhibit the elongation phase of transcription (Ljungman and Hanawalt, 1996; Mello et al., 1995; Sauerbier and Hercules, 1978; Sobell, 1985). Treatments with the kinase inhibitors 5,6-dichloro-1-b-Dribofuranosylbenzimidazole (DRB) or 1-(5-isoquinolinylsulfonyl)-3-methylpiperazine (H7) inhibit the transition from the initiation stage to the elongation stage of transcription by inhibiting phosphorylation of the CTD of RNA polymerase II (Dubois et al., 1994; Marshall et al., 1996). The doses of these dierent agents were chosen so that they would be expected to inhibit mRNA synthesis to a similar degree by 70 ± 90% (Ljungman and Hanawalt, 1996; Ljungman et al., 1999). Using speci®c antibodies recognizing phosphorylation at the ser15 site or acetylation at the lys382 site of p53 we found that UV light, camptothecin and actinomycin D induced modi®cations at both the ser15 and the lys382 sites of p53 (Figure 1). In contrast, DRB and H7 did not induce these p53 modi®cations despite inducing high levels of p53 protein. These results suggest that blockage of
5965
Results Inhibition of elongation, but not promoter escape, leads to phosphorylation of the ser15 and acetylation of the lys382 sites of p53 We have previously shown that inhibition of transcription leads to p53 accumulation in human ®broblasts (Ljungman et al., 1999). The p53 protein is modi®ed at multiple sites following exposure to DNA-damaging agents (Ljungman, 2000), but whether inhibition of transcription results in p53 modi®cations is not known. In this study we investigated whether the accumulation of p53 following inhibition of transcription was accompanied by speci®c modi®cations at the ser15 and lys382 sites of p53. We chose to analyse phosphorylation of ser15 because this modi®cation has been shown to lead to increased stability of p53 (Shieh et al., 1997; Siliciano et al., 1997) and to increase transactivation by binding to the transcription co-factors p300/CBP (Dumaz and Meek, 1999; Lambert et al., 1998). Furthermore, the acetylation of lys382 is associated with increased stability and sequence-speci®c DNA binding activity of p53 (Gu and Roeder, 1997; Ito et al., 2001; Yuan et al., 1999). UV light (254 nm) induces primarily bulky DNA lesions (Friedberg et al., 1995), camptothecin is a DNA topoisomerase I inhibitor that traps the topoisomerase
Figure 1 (a) Induction of ser15 modi®cations of p53 by agents that block the elongation of RNA polymerase II. Human diploid ®broblasts were exposed to 20 J/m2 of UV light (254 nm), 20 mM camptothecin, 200 nM actinomycin D, 100 mM DRB or 100 mM H7. After 2 h, cells were harvested and the levels of ser15-modi®ed p53 (top lanes) and total amount of p53 (lower lanes) were analysed by Western blot using ser15 phosphorylation-speci®c and anti-p53 antibodies. (b) Induction of lys382 acetylation by agents that block the elongation of RNA polymerase II. Experiments were performed as in (a) but cells were incubated for 24 h before harvest. Western blot analysis was performed using lys382 acetylation-speci®c and anti-p53 antibodies Oncogene
Transcription blockage and p53 signaling M Ljungman et al
5966
elongation by DNA lesions induced by UV light, camptothecin or actinomycin D may elicit a unique stress signaling pathway inducing modi®cations at the ser15 the lys382 sites of p53. DRB and H7, on the other hand, abolish transcription by inhibiting the transition from initiation to elongation (Dubois et al., 1994; Marshall et al., 1996), a scenario leading to accumulation of p53 proteins without the induction of modi®cations at ser15 or lys382. We next investigated the time course for the induction of phosphorylation of the ser15 and lys382 sites of p53. In these experiments we pretreated the cells with the proteasome and calpain I inhibitor Nacetyl-L-leucinyl-L-leucinyl-L-norleucinal (LLnL) to accumulate p53 to equivalent levels in all cell samples. The cells were then either treated with actinomycin D, DRB or UV light and incubated for dierent periods of time. It was found that treatment of cells with actinomycin D resulted in the induction of ser15 phosphorylation within 60 min (Figure 2a). The ®nding that actinomycin D induced phosphorylation of the ser15 site in human ®broblasts (Figures 1a and 2a) is in contrast with a previous report (Ashcroft et al., 2000). This discrepancy may be due to the dierent doses of actinomycin D used (5 mM in the other study) or the cell types that were used. UV light induced ser15 modi®cations within 15 min (Figure 2b). Induction of
Figure 2 Time courses for the induction of ser15 and lys382 modi®cations of p53. Cells were pre-incubated for 4 h with the proteasome inhibitor LLnL (50 mM) to induce similar levels of p53 in all cells. (a) Human ®broblasts were then incubated in the presence of 200 nM actinomycin D for dierent periods. Ser15 phosphorylation was assessed as in Figure 1. (b) Rapid induction of ser15 phosphorylation after UV-irradiation (20 J/m2) in HCT116 cells. Ser15 phosphorylation and total p53 levels were determined as described above. (c) Human ®broblasts were treated with 20 J/m2, 100 mM DRB or 200 nM actinomycin D and the cells were incubated for 6 or 24 h. Lys382 acetylation was then assessed as described in Figure 1b Oncogene
lys382 acetylation was observed following 6 h of actinomycin D treatment and 24 h after UV-irradiation but was not detected after either 6 or 24 h of DRB treatment (Figure 2c). Is the lack of ser15 phosphorylation following incubation with the CTD kinase inhibitors DRB and H7 due to inhibition of ser15 kinase(s)? To test this hypothesis we asked whether DRB treatment would inhibit ser15 phosphorylation induced by UV light or ionizing radiation. It was found that post-incubation with DRB did not inhibit the ser15 phosphorylation induced by either UV light or ionizing radiation (Figure 3c). Thus, the lack of phosphorylation of the p53 proteins accumulating within the ®rst 4 h of DRB treatment was not due to an inhibitory eect of DRB on the kinase(s) phosphorylating the ser15 site of p53. Furthermore, time course experiments with DRB or H7 revealed that although the p53 proteins that accumulated within 2 ± 4 h of treatment were not modi®ed, incubations for 8 and 16 h resulted in the phosphorylation of the ser15 site of p53 (Figure 3b,c). This suggests that prolonged incubation with these RNA synthesis inhibitors may induce a secondary
Figure 3 DRB does not inhibit p53 ser15 kinase(s). (a) Human ®broblasts pretreated for 4 h with the proteasome inhibitor MG132 (10 mM) to accumulate p53 to similar levels, were irradiated with 5 Gy of ionizing radiation or 20 J/m2 of UV irradiation and post-incubated in the absence or presence of 100 mM DRB for 2 h. As can be seen, DRB did not interfere with IR- or UV-induced phosphorylation of the ser15 site of p53. (b) Human ®broblasts were incubated with either 100 mM DRB or (c) with 100 mM H7 for dierent periods of time before being harvested and analysed for ser15 phosphorylation as described above. As a negative control for ser15 phosphorylation we included cells treated with the proteasome inhibitor MG132 (10 mM for 4 h). The results show that ser15 phosphorylation of p53 appears even in DRB or H7-treated cells although the initial p53 accumulation is not accompanied by this modi®cation
Transcription blockage and p53 signaling M Ljungman et al
stress response leading to the phosphorylation of the ser15 site of p53. However, it is clear that the initial accumulation of p53 induced by these inhibitors does not require ser15 phosphorylation. Taken, together, our results suggest that DRB or H7 do not inhibit the ser15 kinase(s). Accumulation of p53 in the nucleus following inhibition of transcription We next analysed the cellular localization of p53 proteins accumulating after inhibition of transcription and whether this localization is dependent on the phosphorylation status of the ser15 site. Using immunocytochemistry we show that exposure of human ®broblasts to either UV light, camptothecin, actinomycin D, DRB or H7 resulted in a strong accumulation of p53 in the nucleus within 4 h (Figure 4, top row). Since at the doses used in this experiment these agents have in common that they inhibit mRNA synthesis with about 70 ± 90%, these results suggest that inhibition of transcription is responsible for the nuclear accumulation of p53. We also stained ®xed cells with the phospho-speci®c ser15 antibodies and found that only cells treated with UV light, camptothecin or actinomycin D exhibited any detectable staining (Figure 4, bottom row). These results con®rm the results obtained using Western blots (Figures 1a and 3a,b) in that following 2 or 4 h treatments with DRB or H7, p53 accumulated in the cells without any evidence for concomitant ser15 phosphorylation. Taken together, the results suggest that inhibition of transcription is sucient for the induction of nuclear accumulation of p53. Furthermore, phosphorylation of ser15 is not required for either the accumulation or nuclear localization of p53 following inhibition of transcription. Phosphorylation of the ser15 and acetylation of the lys382 site of p53 following UV-irradiation is triggered by persistent damage in the transcribed strand of active genes UV light, camptothecin and actinomycin D have very dierent mechanisms of action and may induce stress
responses unrelated to inhibition of transcription. To more speci®cally establish that inhibition of transcription is responsible for inducing a stress response resulting in p53 modi®cations, we took advantage of a genetic approach utilizing cells with speci®c DNA repair defects. These cells included normal human ®broblasts (NF) which are pro®cient in global genomic repair (GGR) as well as in transcription-coupled repair (TCR), CS-A ®broblasts which are de®cient in TCR, XP-C ®broblasts which are de®cient in GGR, and XPA ®broblasts which are de®cient in both of these repair pathways. Using these cells we addressed whether UVlight, which is known to aect multiple targets in cells (Bender et al., 1997), would induce p53 modi®cations at ser15 and lys382 by triggering membrane-mediated events or by inducing DNA damage and subsequent blockage of transcription. It was found that ser15 phosphorylation was induced in all four cell lines 2 h after irradiation with a dose as low as 1 J/m2 of UV light in cells pretreated with the proteasome inhibitor MG132 (Figure 5). This dose is below the dose required to induce the accumulation of the p53 protein in human ®broblasts (Hupp et al., 1995; Ljungman and Zhang, 1996; Yamaizumi and Sugano, 1994) and thus could only be detected in cells in which p53 had been accumulated by pretreatment with MG132. The ser15 phosphorylation was found to be transient in normal and XP-C cells with a peak at 2 h and signi®cantly reduced phosphorylation within 4 ± 8 h (Figure 5a). It is expected that the induced photolesions would be eciently removed from the transcribed strand at this time in these TCR-pro®cient cells (Mellon et al., 1987; Venema et al., 1991). In contrast, ser15 phosphorylation was sustained in XP-A cells and increased after longer incubation times in CSA cells. Furthermore, acetylation of the lys382 site of p53 was induced in XP-A and CS-A cells but not in NF and XP-C cells (Figure 5b). These results demonstrate that the trigger for the phosphorylation of the ser15 site and acetylation of the lys382 site of p53 following UV-irradiation was due to DNA damage. Furthermore, since cells with pro®cient TCR lost the ser15 phosphorylation within 4 ± 8 h and
5967
Figure 4 Exposure of human ®broblasts to agents that negatively aect mRNA synthesis results in the nuclear accumulation of p53. Cells were irradiated with 20 J/m2 of UV light (254 nm), 20 mM camptothecin, 200 nM actinomycin D, 100 mM DRB or 100 mM H7. After 4 h the cells were ®xed and p53 localization was analysed using anti-p53 or ser15 phosphorylation-speci®c antibodies and ¯uorescent microscopy. The top panels show the p53 localization while the bottom panels show the localization of ser15-modi®ed p53 proteins Oncogene
Transcription blockage and p53 signaling M Ljungman et al
5968
Figure 5 Cells with de®cient TCR (XPA and CS-A) trigger a sustained ser15 phosphorylation and induce lys382 acetylation of p53 following exposure to 1 J/m2 of UV light (254 nM). Phosphorylation of ser15 (a) and acetylation of lys382 (b) following UVirradiation were assessed in human diploid ®broblasts from normal (NF), XP-A, CS-A and XP-C individuals. Cells were pretreated with 10 mM MG132 for 4 h before being irradiated with 1 J/m2 of UV light (254 nm). After dierent periods of time, the cells were harvested and the speci®c phosphorylation of the ser15 site or acetylation of the lys382 site of p53 were assessed as described in Figure 1
showed no detectable induction of acetylation at the lys382 site, these results suggest that lesions speci®cally in the transcribed strand of active genes trigger a stress response leading to ser15 and lys382 modi®cations of p53. Discussion We have previously shown that inhibition of RNA synthesis is sucient to induce p53 accumulation in human diploid ®broblasts (Ljungman et al., 1999). In this study, we show that the p53 proteins accumulating following treatment with agents that inhibit transcription elongation were phosphorylated at the ser15 site and acetylated at the lys382 site. In contrast, agents that block the transition from initiation to elongation by inhibiting phosphorylation of the CTD of RNA polymerase II induced the accumulation of p53 proteins that were not modi®ed at the ser15 or lys382 sites. Furthermore, the p53 proteins accumulating after inhibition of transcription were localized to the nucleus regardless of whether they were modi®ed at ser15 or not. Finally we show that the induction of ser15 and lys382 modi®cations by UV light is speci®cally triggered by Oncogene
UV-induced lesions in the transcribed strand of active genes. Taken together, our results strongly suggest that a general inhibition of transcription is sucient to induce accumulation of p53 in the nucleus while modi®cations of p53 are speci®cally triggered by blockage of elongating RNA polymerase II complexes (Figure 6). Phosphorylation of the N-terminus of p53, including ser15, is thought to increase the stability of the p53 protein by attenuating the interaction with its negative regulator MDM2 (Shieh et al., 1997; Siliciano et al., 1997). However, we found that p53 proteins accumulating in the nucleus following treatment with DRB or H7 were not accompanied by phosphorylation at the ser15 site. Thus, nuclear accumulation of p53 does not require phosphorylation of ser15 of p53 which support a previous report (Blattner et al., 1999b). It is possible that transcription inhibition leads to nuclear accumulation of p53 by an indirect mechanism involving the loss of MDM2 expression (Ashcroft et al., 2000; Blattner et al., 1999a) or by interference with the nuclear export machinery (Groulx et al., 2000). Interestingly, UV light, camptothecin and actinomycin D, which at the doses used inhibit the elongation phase of transcription, rapidly induced phosphorylation of the ser15 site of p53. However, since these agents also directly aect
Transcription blockage and p53 signaling M Ljungman et al
Figure 6 The elongating RNA polymerase II act as a sensor of DNA damage and an integrator of stress signals leading to p53 modi®cations. Inhibition of CTD kinases by H7 and DRB leads to inhibition of RNA synthesis and nuclear accumulation of p53 proteins but does not lead to modi®cations of ser15 or lys382. UV light (254 nm) induces primarily pyrimidine dimers, actinomycin D intercalates DNA, and camptothecin traps DNA topoisomerase I in a cleavable complex bound to DNA. These lesions block the elongation step of transcription and result in the modi®cation of p53 at ser15 and lys382. The triggering mechanism leading to the activation of the p53-modifying enzymes may involve the stalled RNA polymerase itself or associated transcription-, elongation- or RNA processing-factors associated with the phosphorylated CTD of the elongating RNA polymerase II
the structure of DNA, it is possible that the stress response leading to p53 modi®cations is triggered by DNA alterations rather than by blockage of transcription elongation. To distinguish whether the phosphorylation of the ser15 site of p53 following UV-irradiation was due to blockage of transcription elongation or some alternative mechanism, we took advantage of cell lines with speci®c defects in DNA repair. UV-irradiation is known to trigger multiple signal transduction pathways in cells (Bender et al., 1997). Some of these pathways are triggered from the cell membrane, while others are triggered by DNA damage (Bender et al., 1997). In this study we show that the phosphorylation of the ser15 site and the acetylation of the lys382 site of p53 following irradiation with 1 J/m2 of UV light were induced speci®cally by persistent DNA damage. This was deduced from experiments comparing the persistence of ser15 phosphorylation following UV irradiation in DNA repair pro®cient and de®cient cells. While ser15 phosphorylation of p53 in the nucleotide excision repair-defective XP-A cells persisted for more than 8 h, this modi®cation was transient in normal ®broblasts (Figure 5). Furthermore, induction of lys382 acetylation was observed 24 h after UV-irradiation in XP-A but not in normal cells. Importantly, we found that XP-C cells resembled normal cells in that they showed a transient induction of ser15 and no induction of acetylation of lys382, while CS-A cells resemble XP-A cells in that UV-irradiation induced a persistent phosphorylation of ser15 and induction of acetylation of lys382 by 24 h. These results suggest that the induction of ser15 phosphorylation and lys382 acetylation following UV-irradiation is triggered speci®cally by lesions in the transcribed strand of active genes and strongly implicate blockage of elongation in the triggering of these p53 modi®cations.
The kinetics of lys382 induction following UVirradiation appears to be much slower than the induction of the ser15 phosphorylation or the overall accumulation of the p53 protein in the UV-irradiated cells. Thus, from our data we can not conclude that the lys382 modi®cation is important for the accumulation of p53. A recent study, however, found that the rate of induction of lys382 phosphorylation following exposure of cells to UV light or actinomycin D closely followed the rate of p53 accumulation (Ito et al., 2001). Importantly, our results using the DNA repair-de®cient cells show that persistent DNA lesions in the transcribed strand of active genes are responsible for the triggering of the lys382 modi®cation after UVirradiation. Thus, although perhaps not involved in the accumulation of p53 in these cells after UV-irradiation, our results strongly suggest the lys382 modi®cation is induced by a pathway or stress response that is linked to blocked RNA polymerase II complexes. One important ®nding of our study was that DRB and H7 did not induce phosphorylation of ser15 or acetylation of lys382 (Figure 1). These agents inhibit RNA polymerase II-mediated transcription while they are not interfering with the synthesis of rRNA by RNA polymerase I (Dubois et al., 1994; Marshall et al., 1996). However, UV light, camptothecin and actinomycin D do aect RNA polymerase I-mediated transcription and thus, it is possible that the modi®cations of p53 induced by these agents are due to their inhibitory action of RNA polymerase I elongation. However, our results with XP-C cells, which have pro®cient TCR of RNA polymerase II-transcribed genes but de®cient repair of ribosomal RNA genes following UV-irradiation (Christians and Hanawalt, 1994), showed that these cells did not induce a persistent ser15 phosphorylation nor did they induce acetylation of lys382 following UV-irradiation. Thus, these results argue against the requirement of blockage of RNA polymerase I elongation for the induction of p53 modi®cations. What is the mechanism by which blockage of the elongation of RNA polymerase II complexes triggers p53 modi®cations? Following initiation of transcription, the carboxyl-terminal domain (CTD) of RNA polymerase II becomes phosphorylated at multiple sites ®rst by the Cdk7 subunit of TFIIH and later by the Cdk9 subunit of P-TEFb (Price, 2000). The phosphorylated CTD is thought to be an assembly point for enzymes involved in RNA capping, splicing and polyadenylation (Hirose and Manley, 2000; Proudfoot, 2000). Blockage of RNA synthesis by DNA lesions in the transcribed strand would presumably halt these RNA processing activities. Thus, the signal triggered by the stalled elongation complex leading to p53 modi®cations may be induced by transcription factors or RNA processing factors associated with the elongating RNA polymerase (Figure 5). The processing and degradation of the largest subunit of the blocked RNA polymerase II is less likely to trigger these stress signals since we recently showed that Cockayne's syndrome cells are defective in this degradation
5969
Oncogene
Transcription blockage and p53 signaling M Ljungman et al
5970
(McKay et al., 2001) but yet in this study, we show that CS cells are pro®cient in the modi®cation of both the ser15 and lys382 sites following UV-irradiation. The reason that blockage of RNA polymerase at promoter-proximal regions does not trigger a signaling pathway leading to p53 modi®cations may be that this type of stalling occurs naturally following initiation of most genes before the CTD of the RNA polymerase becomes phosphorylated (Price, 2000). In contrast, blockage at promoter-distal sites may be recognized as inappropriate and results in the mounting of a distinct stress response pathway resulting in p53 modi®cations. Which are the p53-modifying enzymes that respond to transcriptional stress? The ATR kinase has been shown to be the principal p53 ser15 kinase following UV-irradiation (Tibbetts et al., 1999). Since our study using DNA repair-de®cient human cells show that the induction of ser15 phosphorylation is closely linked to persistent DNA lesions speci®cally in the transcribed strand, it is plausible that the ATR kinase may be a link between blocked transcription and p53. Furthermore, the acetylation of the lys382 site is known to be carried out by the acetyl transferase p300 (Sakaguchi et al., 1998). Since p300 has been shown to be associated with the RNA polymerase II holoenzyme (Neish et al., 1998), it could potentially have the ability to sense stalling of the RNA polymerase II complex and signaling to p53. Future studies will be directed to elucidate the roles, if any, of the ATR kinase and the p300 acetyltransferase in signaling p53 following blockage of transcription. In conclusion, we show that the accumulation of p53 in the nucleus following inhibition of transcription is not strictly dependent on phosphorylation of the ser15 site of p53. The loss of MDM2 expression may play a role in this nuclear accumulation (Ashcroft et al., 2000; Blattner et al., 1999a) but other mechanisms such as interference with the nuclear export machinery may also contribute (Groulx et al., 2000). Furthermore, our results strongly suggest that the modi®cations of the ser15 and the lys382 sites of p53 following UVirradiation are generated by an active mechanism triggered by the blockage of elongating RNA polymerase II complexes. Previous studies have shown on the connection between blocked transcription and DNA repair (Le Page et al., 2000; Mellon et al., 1987), p53 (Ljungman and Zhang, 1996; Ljungman et al., 1999; Yamaizumi and Sugano, 1994) and apoptosis (Ljungman and Zhang, 1996). The results from the present study further demonstrate the central role that the elongating transcription machinery play as a general sensor of DNA damage and as an integrator of DNA damage stress signals.
Materials and methods Cell culture Diploid human ®broblasts (NF, XP-A, XP-C, CS-A) or the human colon cancer cell line HCT116 were grown as Oncogene
monolayers on culture dishes or on coverslips in MEM or RPMI (HCT116) supplemented with 10% fetal bovine serum (FBS) and antibiotic/antimycotic (GIBCO BRL). Western blots Following treatment, cells were rinsed with PBS, detached by scraping and collected by centrifugation. Protein concentration was quanti®ed using a protein assay (BioRad) and approximately 30 mg of protein was loaded per lane. The cell samples were boiled for 5 min in loading buer (2% SDS, 10% glycerol, 5% 2-mercaptoethanol, 0.05% bromophenol blue and 62.5 mM Tris pH 6.8) prior to loading. Following SDS ± PAGE, proteins were electrophoretically transferred to Immobilon-P transfer membranes (Millipore). Antibodies used for the Western blot were anti-ser15 phospho-speci®c antibody (Ser15, Ab-3, New England Biolabs) anti-lys382 acetylation-speci®c antibody (Lys382, Oncogene Research Products) and anti-p53 antibody (Ab-2, Oncogene Research Products). X-ray ®lm (Biomax AR, Amersham) and enhanced chemiluminescence (Super Signal CL-HRP Substrate System, Pierce) were used to visualize the p53 proteins. The quality of total protein transfer was assessed by staining blots with Coomassie brilliant blue following exposure of the membranes to X-ray ®lm. Immunocytochemistry Human ®broblasts grown on coverslips were irradiated or treated with various agents for 4 h. The cells were then washed with PBS and ®xed in ice-cold methanol/acetone (1 : 1) and stored at 7208C for 1 h. The ®xed cells were rinsed twice in PBS and incubated for 1 h with the mouse monoclonal anti-p53 antibody 1801 (a gift from Dr Jiayuh Lin, University of Michigan) or the anti-ser15 phosphospeci®c antibody (Ser15, Ab-3, New England Biolabs). The antibody solution was aspirated and the cells were rinsed on a rocker platform three times for 5 min with PBSBT (5 g bovine albumin and 500 ml tween-20 per liter of PBS). The ®xed cells were then incubated for 1 h in the dark with 100 ml of a secondary FITC-conjugated anti-mouse IgG antibody (Sigma; 1 : 1000 dilution in PBSBT). Following rinsing in PBSBT as described above, the coverslips were mounted on microscope slides in one drop of Vectashield (Vector Laboratories, Inc.). Images were captured digitally with Adobe Photoshop software using a Zeiss Akioskop ¯uorescent microscope with Plan-NEUFLUAR 636/1.25 oil lens and a Microimage i308 digital camera. The color images were converted to grayscale.
Acknowledgments We thank Bruce McKay for valuable input and Sara Ljungman for critically reading this manuscript. This work was supported by a grant from the National Institute of Health (CA82376-01).
Transcription blockage and p53 signaling M Ljungman et al
5971
References An WG, Chuman Y, Fojo T and Blagosklonny MV. (1998). Exp. Cell. Res., 244, 54 ± 60. Ashcroft M, Taya Y and Vousden KH. (2000). Mol. Cell. Biol., 20, 3224 ± 3233. Banin S, Moyal L, Shieh SY, Taya Y, Anderson CW, Chessa L, Smorodinsky NI, Prives C, Reiss Y, Shiloh Y and Ziv Y. (1998). Science, 281, 1674 ± 1677. Bender K, Blattner C, Knebel A, lordanov M, Herrlich P and Rahmsdorf HJ. (1997). J. Photochem. Photobiol. B: Biol., 37, 1 ± 17. Blattner C, Sparks A and Lane D. (1999a). Mol. Cell. Biol., 19, 3704 ± 3713. Blattner C, Tobiasch E, Litfen M, Rahmsdorf HJ and Herrlich P. (1999b). Oncogene, 18, 1723 ± 1732. Boyd SD, Tsai KY and Jacks T. (2000). Nat. Cell Biol., 2, 563 ± 568. Canman CE, Lim DS, Cimprich KA, Taya Y, Tamai K, Sakaguchi K, Appella E, Kastan MB and Siliciano JD. (1998). Science, 281, 1677 ± 1679. Chehab NH, Malikzay A, Stavridi ES and Halazonetis TD. (1999). Proc. Natl. Acad. Sci. USA, 96, 13777 ± 13782. Chen AY and Liu LF. (1994). Annu. Rev. Pharmacol. Toxicol., 34, 191 ± 218. Christians FC and Hanawalt PC. (1994). Mutat. Res., 323, 179 ± 187. Dubois MF, Nguyen VT, Bellier S and Bensaude O. (1994). J. Biol. Chem., 269, 13331 ± 13336. Dumaz N and Meek DW. (1999). EMBO J., 18, 7002 ± 7010. Fiscella M, Ullrich S, Zambrano N, Shields M, Lin D, LessMiller S, Anderson C, Mercer W and Apella E. (1995). Oncogene, 8, 1519 ± 1528. Friedberg E, Walker G and Siede W. (1995). DNA repair and mutagenesis. ASM Press: Washington, D.C. Geyer RK, Yu ZK and Maki CG. (2000). Nat. Cell Biol., 2, 569 ± 573. Groulx I, Bonicalzi ME and Lee S. (2000). J. Biol. Chem., 275, 8991 ± 9000. Gu W and Roeder RG. (1997). Cell, 90, 595 ± 606. Hirose Y and Manley JL. (2000). Genes Dev., 14, 1415 ± 1429. Hupp T, Sparks A and Lane D. (1995). Cell, 83, 237 ± 245. Ito A, Lai CH, Zhao X, Saito S, Hamilton MH, Appella E and Yao TP. (2001). EMBO J., 20, 1331 ± 1340. Koumenis C and Giaccia A. (1997). Mol. Cell. Biol., 17, 7306 ± 7316. Lambert PF, Kashanchi F, Radonovich MF, Shiekhattar R and Brady JN. (1998). J. Biol. Chem., 273, 33048 ± 33053. Le Page F, Kwoh EE, Avrutskaya A, Gentil A, Leadon SA, Sarasin A and Cooper PK. (2000). Cell, 101, 159 ± 171. Ljungman M. (2000). Neoplasia, 2, 208 ± 225. Ljungman M and Hanawalt PC. (1996). Carcinogenesis, 17, 31 ± 35. Ljungman M and Zhang F. (1996). Oncogene, 13, 823 ± 831. Ljungman M, Zhang FF, Chen F, Rainbow AJ and McKay BC. (1999). Oncogene, 18, 583 ± 592. Lu X, Burbidge SA, Grin S and Smith HM. (1996). Oncogene, 13, 413 ± 418.
Marshall NF, Peng JM, Xie Z and Price DH. (1996). J. Biol. Chem., 271, 27176 ± 27183. McKay B, Chen F, Clarke S, Wiggin H, Harley L and Ljungman M. (2001). Mutat. Res., 485, 93 ± 105. McKay B and Ljungman M. (1999). Neoplasia, 1, 276 ± 284. McKay BC, Chen F, Perumalswami CR, Zhang FF and Ljungman M. (2000). Mol. Biol. Cell, 11, 2543 ± 2551. McKay BC, Ljungman M and Rainbow AJ. (1998). Oncogene, 17, 545 ± 555. Mello JA, Lippard SJ and Essigmann JM. (1995). Biochemistry, 34, 14783 ± 14791. Mellon I, Spivak G and Hanawalt PC. (1987). Cell, 51, 241 ± 249. Neish AS, Anderson SF, Schlegel BP, Wei WJ and Parvin JD. (1998). Nucleic Acids Res, 26, 847 ± 853. Perry M, Piette J, Zawadzki J, Harvey D and Levine A. (1993). Proc. Natl. Acad. Sci. USA, 90, 11623 ± 11627. Poele RHT, Okorokov AL and Joel SP. (1999). Oncogene, 18, 5765 ± 5772. Price DH. (2000). Mol. Cell. Biol., 20, 2629 ± 2634. Proudfoot N. (2000). Trends Biochem. Sci., 25, 290 ± 293. Reinke V and Lozano G. (1997). Radiat. Res., 148, 115 ± 122. Sakaguchi K, Herrera JE, Saito S, Miki T, Bustin M, Vassilev A, Anderson CW and Appella E. (1998). Genes Dev., 12, 2831 ± 2841. Sakaguchi K, Saito S, Higashimoto Y, Roy S, Anderson CW and Appella E. (2000). J. Biol. Chem., 275, 9278 ± 9283. Sauerbier W and Hercules K. (1978). Annu. Rev. Genet., 12, 329 ± 363. She QB, Chen N and Dong Z. (2000). J. Biol. Chem., 275, 20444 ± 20449. Shieh SY, Ikeda M, Taya Y and Prives C. (1997). Cell, 91, 325 ± 334. Siliciano JD, Canman CE, Taya Y, Sakaguchi K, Appella E and Kastan MB. (1997). Genes Dev., 11, 3471 ± 3481. Sobell H. (1985). Proc. Natl. Acad. Sci. USA, 82, 5328 ± 5331. Tibbetts RS, Brumbaugh KM, Williams JM, Sarkaria JN, Cliby WA, Shieh SY, Taya Y, Prives C and Abraham RT. (1999). Genes Dev., 13, 152 ± 157. Unger T, Juven-Gershon T, Moallem E, Berger M, Sionov RV, Lozano G, Oren M and Haupt Y. (1999a). EMBO J., 18, 1805 ± 1814. Unger T, Sionov RV, Moallem E, Yee CL, Howley PM, Oren M and Haupt Y. (1999b). Oncogene, 18, 3205 ± 3212. Venema J, Vanhoen A, Karcagi V, Natarajan AT, Vanzeeland AA and Mullenders LHF. (1991). Mol. Cell. Biol., 11, 4128 ± 4134. Wang JA, Fan S, Yuan RQ, Ma YX, Meng Q, Goldberg ID and Rosen EM. (1999). Int. J. Radiat. Biol., 75, 301 ± 316. Yamaizumi M and Sugano T. (1994). Oncogene, 9, 2775 ± 2784. Yuan ZM, Huang YY, Ishiko T, Nakada S, Utsugisawa T, Shioya H, Utsugisawa Y, Yokoyama K, Weichselbaum R, Shi Y and Kufe D. (1999). J. Biol. Chem., 274, 1883 ± 1886.
Oncogene
