DN-p73 is activated after DNA damage in a p53-dependent manner to regulate p53-induced cell cycle arrest

Article metrics

Abstract

p53 and p73 genes are both activated in response to DNA damage to induce either cell cycle arrest or apoptosis, depending on the strength and the quality of the damaging stimulus. p53/p73 transcriptional activity must be tightly regulated to ensure that the appropriate biological response is achieved and to allow the cell to re-enter into the cell cycle after the damage has been repaired. In addition to multiple transcriptionally active (TA) isoforms, dominant negative (DN) variants, that lack the amino-terminal transactivation domain and function as trans-repressors of p53, p63 and p73, are expressed from a second internal promoter (P2-p73Pr). Here we show that, in response to a non apoptotic DNA damage induced by low doses of doxorubicin, p53 binds in vivo, as detected by a p53-specific chromatin immunoprecipitation assay, and activates the P2-p73 promoter. DN-p73α protein accumulates under the same conditions and exogenously expressed DN-p73α is able to counteract the p53-induced activation of the P2-p73Pr. These results suggest that DN-p73 may contribute to the autoregulatory loops responsible for the termination of p53/p73 responses in cells that do not undergo apoptosis. Accordingly, the activation of the P2-p73Pr is markedly enhanced in both p73−/− murine fibroblasts and in human cells in which p73 transcripts are selectively knocked-out by p73-specific small interfering RNAs.

Damage to chromosomal DNA induces a complex cellular response designed to delay cell cycle progression until the DNA is repaired (Zhou and Elledge, 2000). In the event of irreparable damage the cell undergoes apoptotic cell death in order to preserve genome integrity. Response to DNA damage is critical for cell viability. Unchecked DNA damage can lead to mutations, translocations and abnormal recombination events during S phase and also chromosome breakage and loss during mitosis. Failure to monitor or signal damaged DNA is a hallmark of cancer cells (Hartwell and Kastan, 1994). The choice between cell cycle arrest and apoptosis is influenced by both the quality and the strength of the DNA damage imposed to the cell. In general, cell cycle arrest is prevalent when cells are treated with low concentrations of DNA damaging agents and apoptosis is triggered in response to higher dosages. Several signal transduction pathways are activated and integrated within the cell to implement the appropriate decision (Rich et al., 2000). Many of these signals converge to regulate the activity of both the p53 tumor suppressor gene (Prives and Hall, 1999) and its family members p63 or p73 (Agami et al., 1999; Gong et al., 1999; Levrero et al., 1999, 2000; Yuan et al., 1999). Differently from p53, p73 exists as a group of full length isoforms that arise from alternative splicings at the C-terminus (TA-p73 α-ζ) (De Laurenzi et al., 1998, 1999) and are all able to induce both cell cycle arrest and apoptosis when overexpressed. In addition, amino-terminally truncated isoforms (DN-p73s) that lack the transactivation domain and exert a dominant negative function towards p53, p73 and p63 activity have been described (Sayan et al., 2001). DN-p73 is expressed in vivo in the developing brain where it counteracts p53 dependent apoptosis (Pozniak et al., 2000). Similarly, truncated isoforms of the related protein p63 may act as dominant negative inhibitors of p53 and p73 dependent transcription (Yang et al., 1998) and protect keratinocytes from DNA damage induced apoptosis in vitro and in vivo (Liefer et al., 2000). Full-length TA-p73 and truncated DN-p73 mRNAs are transcribed from two different promoters named P1-p73Pr and P2-p73Pr, respectively. Whereas the regulation of the P1-p73 promoter has been characterized in several cellular systems and found to be activated in a E2F1-dependent manner (Irwin et al., 2000; Lissy et al., 2000; Pediconi et al., manuscript in preparation) very little is known about the transcriptional regulation of the DN-p73 proteins. To get insights into how the P2-p73Pr is regulated by extracellular stimuli we isolated a 2.4 Kb region of the p73 gene placed upstream the putative DN-p73 transcription initiation site and containing the putative P2-p73 promoter. Sequence analysis and binding sites scanning by the MacVector software (Kodak, Inc.) revealed the presence of several putative transcription factor binding sites, including two sequences bearing substantial homology to consensus p53 responsive elements (p53RE) between position −75 and −56 relative to the first nucleotide of the DN-p73 specific exon 3′ (Figure 1a). To assess whether the P2-p73 promoter may be indeed regulated by p53 we subcloned the 2.4 Kb genomic region into the pGL3-basic plasmid (Promega, Inc.) to obtain the pGL3-P2-p73 (-2400) luciferase reporter and performed transient cotransfection assays in the p53 negative SAOS-2 osteosarcoma cell line. As predicted, P2-p73-luc transcription is strongly upregulated by both overexpressed p53 and p73α (Figure 1b), whereas neither a mutated p73 that is not able to bind DNA (p73A154V) nor DN-p73α are able to induce transcription from the same reporter, thus indicating that both the DNA binding and the presence of the transactivation domain are required to activate P2-p73Pr transcription (Figure 1b). Truncation of the putative p53RE abolished the p53-induced transcription of the P2-p73 promoter (Figure 1c). To confirm the in vivo relevance of this finding we investigated the ability of transcription competent p53 and p73 proteins to activate the endogenous P2-p73 promoter. To this aim total RNAs were prepared from mock transfected cells (pCDNA) and from cells transfected with plasmid vectors expressing p53, p73α, p73β, DN-p73α and the p73αA154V mutant and subjected to semiquantitative RT–PCR. As shown in Figure 1d, exogenously expressed p53 and p73β, and to a lesser extent p73α, activate transcription from the endogenous P2-p73 promoter.

Figure 1
figure4

The P2-p73 promoter is upregulated by the p53 family members. (a) Schematic representation of the 5′ region of the P73 locus. P1 and P2 indicate the position of the promoters for TA-p73 and DN-p73 isoforms, respectively. A 2.4 Kb genomic region located upstream the putative transcription initiation site of DN-p73 mRNA was amplified using specific primers and cloned into the SmaI site of the luciferase reporter plasmid pGL3-basic (Promega, Inc.). Sequence analysis confirmed the identity with the sequence present in Genebank. The indicated deletion mutants of the P2-p73 promoter were generated by standard PCR and cloning techniques. (b) SAOS-2 cells were transfected with 0.5 μg of the P2-p73 firefly luciferase reporter along with 1 μg of the indicated expression vectors and 100 ng of the pRL-null vector, that encodes for the renilla luciferase, using the calcium phosphate method. Twenty-four hours after transfection cells were lysed and assayed for luciferase activity using the dual luciferase assay system (Promega, Inc.). Firefly luciferase activity was normalized to the renilla luciferase activity and expressed as fold activation relative to the level obtained after cotransfection of the reporter vector with the empty expression vector. Error bars represent two standard deviations calculated on three independent experiments. The levels of the exogenously expressed proteins were revealed by immunoblotting using an anti-HA polyclonal antibody (Santa Cruz, Inc.) and are shown in the upper panel. (c) In a similar set of experiments the reporter plasmids carrying the full length P2-p73 promoter and the deletion mutants described in (a) were transfected into SAOS-2 cells in the presence or absence of cotransfected HA-p53 encoding plasmid. Luciferase activity was determined as described in (b). (d) The endogenous P2-p73 promoter is transcriptionally upregulated by the p53 family members. Total RNAs extracted from SAOS-2 cells transfected with the indicated expression plasmids were reverse transcribed using the Single Strand cDNA Synthesis kit (Amersham Pharmacia, Inc.) and the resulting cDNA were amplified with primers specific for DN-p73 and β-Actin. Primer sequences are available upon request

Given the high homology shared by p53, TA-p73s and DN-p73s in their DNA binding domain and their ability to activate largely overlapping sets of genes (Levrero et al., 2000), our findings are consistent with a model in which DN-p73 proteins may be upregulated at the transcriptional level by stimuli that activate the p53 pathway and may act to repress p53-dependent transcription by competing for its binding site on different promoters. To verify this hypothesis we treated the human osteosarcoma U2OS and SAOS-2 cells with doxorubicin, a DNA damaging agent known to activate both p53 and p73 (Gong et al., 1999; Costanzo et al., 2002). As shown in Figure 2a, doxorubicin treatment activates the P2-p73 promoter in the p53 wild type U2OS cell line, indicating that DN-p73 isoforms are a physiological target of the DNA damage response. Interestingly, the induction of the P2-p73 promoter is higher when cells are challenged with a non apoptotic dosage of doxorubicin, whilst a lower level of induction is obtained at a higher concentration of the drug (3 μM) that is known to induce a strong apoptotic response (Oda et al., 2000; Costanzo et al., 2002). Consistent with this model, treatment of U2OS cells with non apoptotic dosages of doxorubicin efficiently activate transcription from the p21 promoter but not from the p53AIP1 promoter (Figure 2a middle and right panels). Moreover, doxorubicin treatment also results in the induction of endogenous DN-p73 in U2OS cells, as demonstrated by immunoblot using a monoclonal antibody that specifically recognizes the DN-p73 proteins. In agreement with the luciferase reporter experiments the amount of endogenous DN-p73 increases only in cells treated with doxorubicin concentrations that induce cell cycle arrest (0.2 or 1 μM) while higher apoptotic concentrations of doxorubicin are not effective in inducing the DN-p73 (Figure 2b).

Figure 2
figure3

p53-dependent activation of the P2-p73 promoter in response to DNA damage. (a) U2OS cells were transfected with the P2-p73, p21 and p53AIP1 luciferase reporter plasmids (Costanzo et al., 2002), treated with increasing concentrations of doxorubicin for 24 h and assayed for luciferase activity. Firefly luciferase activity was normalized to the renilla luciferase activity and expressed as fold activation relative to the level obtained after cotransfection of the reporter vector with the empty expression vector. Error bars represent two standard deviations calculated on three independent experiments. (b) Extracts from U2OS cells, either untreated or treated with the indicated concentrations of doxorubicin, were subjected to immunoblot using antibodies specific for the DN-p73 isoforms (Imgenex Inc.), p53 (clone DO-1; Santa Cruz, Inc.) and anti-β-actin (Santa Cruz, Inc). (c) Human U2OS and SAOS-2 osteosarcoma cells, p53−/− fibroblasts and p53+/+ fibroblasts were transfected with the P2-p73 luciferase reporter, treated with doxorubicin and assayed for luciferase activity 24 h later. Luciferase activity was determined as described in (a). (d) U2OS cells were treated with doxorubicin for either 12 or 24 h, exposed to 1% formaldehyde for 10′ at room temperature and then subjected to a chromatin immunoprecipitation assay using either the anti-p53 clone DO1 (Santa Cruz, Inc.) or an unrelated antibody. The presence of the P2-p73, p21 and GAPDH promoters in the chromatin recovered after immunoprecipitation was examined by PCR using primers specific either for the regions that contain in the above mentioned promoters the p53 binding sites (P2-p73Pr and p21Pr) or the region immediately upstream the GAPDH promoter TATA box. Primer sequences are available upon request

The induction of the P2-p73 promoter after DNA damage is severely impaired in fibroblasts derived from p53−/− mice and in the p53 negative SAOS-2 cell line (Figure 2c). Since in both cell lines transcriptionally active TA-p73 proteins accumulate after treatment with doxorubicin (Costanzo et al., 2002 and data not shown), being therefore available to activate both the endogenous and the transfected P2-p73 promoter, these observations underlie a pivotal role of p53 in the induction of the P2-p73 promoter in response to DNA damage in vivo. The in vivo recruitment of p53 on the P2-p73 promoter has been confirmed by performing a p53-specific chromatin immunoprecipitation (ChIP) assay in cells treated with 0.5 μM doxorubicin. As shown in Figure 2d, DNA recovered from anti-p53 immunoprecipitation of formaldehyde cross-linked U2OS cells was amplified with primers specific for the P2-p73 but not with primers specific for the GAPDH promoter, thus indicating that p53 does not occupy the P2-p73 promoter in untreated proliferating cells and specifically binds the P2-p73 promoter after DNA damage. Interestingly enough, we were able to detect p53 recruitment on the P2-p73 promoter at 12 h of doxorubicin treatment but not at 24 h (Figure 2d) and when higher doxorubicin concentrations (2 μM) were used (data not shown). These results suggest that the upregulation of the P2-p73 promoter by p53 and the accumulation of DN-p73 proteins may represent an important early component of the cellular response to DNA damage.

To further establish the role of DNp73 proteins in the cellular responses to DNA damage, we evaluated the ability of DNp73s to downregulate p53 activity on its own promoter as well as on other p53-dependent promoters. To this aim we performed parallel co-transfection assays in the p53-deficient SAOS-2 cells using luciferase reporter constructs driven by the regulatory sequences of genes involved in either cell cycle arrest (i.e. p21) or in the induction of apoptosis (i.e. Bax, Pig3, p53AIP1). As shown in Figure 3, the ability of exogenously expressed DN-p73 to downregulate p53-induced transcription is greatly influenced by the target promoter. Indeed, DN-p73 efficiently counteracts p53-induced activation of its own promoter, it is still active in inhibiting the Pig3, Bax and p21 promoters but it is unable to affect the activation of the p53AIP1 promoter. It is noteworthy that the p53AIP1 gene encodes for a mitochondrial protein that causes apoptosis and is upregulated by p53 carrying specific post-translational modifications that are induced only by apoptotic levels of DNA damage (Oda et al., 2000). The same modifications do not seem to be required for p53 to activate the Bax and the Pig3 gene (Oda et al., 2000). Taken together these results would suggest that the physiological role of DNp73 in the response to DNA damage may be to regulate the non apoptotic response to p53 activating agents. This would be achieved by a physical competition with p53 for the binding to the promoters of a class of ‘apoptotic’ genes that can be activated by ‘unmodified’ p53 and by modulating the activation of the cyclin dependent kinase inhibitor p21 to allow cell cycle re-entry after the DNA damage is repaired. Would this hypothesis be true the activity of the P2-p73 promoter should be upregulated if the damage to the DNA is reparable and the accumulated DN-p73 proteins should downregulate the activity of p53 on both the P2-p73 promoter itself and on other p53-dependent promoters. To assess the consistency of this model we analysed the ability of the P2-p73 promoter to be upregulated by DNA damage in p73−/− fibroblasts as compared to their wild type counterpart. As shown in Figure 4a, not only the basal activity of the P2-p73 promoter but also its activation after doxorubicin treatment is significantly higher in fibroblasts derived from p73−/− mice as compared with p73+/+ fibroblasts. To further validate our model and to confirm these data in human cells we sought to selectively downregulate p73 gene transcripts in a p53 and p73 wild type cellular context by using small interfering RNA (siRNA) specific for p73. RNA interference (RNAi) or RNA silencing is the process whereby double-stranded RNA (dsRNA) induces a homology-dependent degradation of cognate mRNA. Recently, small interfering RNAs have been shown to achieve a high degree of specificity with low toxicity also in mammalian cells (Caplen et al., 2001; Elbashir et al., 2001) acting through a degradative chain reaction catalyzed by the activation of a cellular RNA-dependent RNA polymerase (Lipardi et al., 2001; Nishikura et al., 2001; Sijen et al., 2001). Transfection of a siRNA specific for the GFP abolishes the expression of a cotransfected GFP expression plasmid without affecting endogenous p73 expression (Figure 4b). We also found that a p73 specific siRNA strongly downregulates the levels of the endogenous p73 (Figure 4b). The effect of the p73 siRNA is highly specific in that it does not affect neither the levels of GFP expressed from a cotransfected vector (Figure 4b) nor the expression of endogenous p53 (data not shown). We evaluated the effects of p73 specific interference on the activity of the P2-p73 promoter by cotransfection assays in U2OS cells. The selective knock-out of p73 transcripts increases the activation of the P2-p73 promoter induced by both low doses (0.5 μM) of doxorubicin (Figure 4c, left panel) and by cotransfected p53 (Figure 4c). These results further confirm the existence of a negative auto-regulatory loop exerted by DN-p73 protein on its own promoter and establish the role of DN-p73 as a regulator of p53 function in the cell cycle arrest phase of the cellular response to doxorubicin.

Figure 3
figure2

SAOS-2 cells were co-transfected with 100 ng of pRL-null plasmid, 0.5 μg of the indicated reporter construct, 0.5 μg of the HA-p53 expression plasmid and with increasing concentrations of HA-DN-p73α encoding plasmid. Cells were lysed 24 h after transfection and the dual luciferase assay (Promega, Inc.) was performed. Luciferase activity is expressed as a percentage of activation being p53-induced activation set as 100%. Results represent the luciferase activity normalized to the renilla activity. Error bars represent two standard deviations calculated on three independent experiments. The level of exogenously expressed p53 and DN-p73α proteins was checked by immunoblotting using an anti-HA-tag antibody (Santa Cruz, Inc.)

Figure 4
figure1

p73−/− fibroblasts and p73+/+ fibroblasts (kindly provided by JYJ Wang) were transfected with the P2-p73 luciferase reporter, treated with doxorubicin and assayed for luciferase activity 24 h later. Luciferase activity is expressed as fold activation relative to the level obtained in untreated p73+/+ cells after cotransfection of the reporter vector with the empty expression vector. Results represent the luciferase activity normalized to the renilla activity. Error bars represent two standard deviations calculated on three independent experiments. (b) U2OS cells were cotransfected with 100 ng of the GFP encoding plasmid GFPN1 (Clontech, Inc.) and siRNAs specific for either GFP (500 ng) and p73 (500 ng) (Xeragon, Inc.) using the Lipofectamine-Plus reagent (Invitrogen, Inc.), according to the protocol developed by Caplen et al. (2001). The 22 nt GFP specific siRNA was previously described and characterized (Caplen et al., 2001). The p73 specific siRNA was designed to target both TA- and DN-isoforms of p73. The sequences of the siRNAs is available upon request. Cell extracts prepared 24 h after transfection were assayed by immunoblotting for the levels of endogenous p73α (clone 1288 that recognizes both TA- and DN-p73 isoforms, Imgenex, Inc) (upper panel) and of exogenously expressed GFP (Santa Cruz, Inc.) (lower panel). (c) U2OS cells were cotransfected with 0.5 μg of P2-p73 luciferase reporter together with the siRNAs specific for either GFP (500 ng) and p73 (500 ng). Eighteen hours after transfection cells were either mock treated or exposed to 0.2 μM of doxorubicin for additional 24 h (left panel). The same experiment was performed cotransfecting the HA-p53 expression vector together with the P2-p73 luciferase reporter, the p73 and the GFP specific siRNAs (right panel). Luciferase activity was determined as described in (a)

The ability of p53 and the p53 family members p73 and p63 to restrain or kill the cell must be tightly controlled under normal conditions. Regulation of p53 occurs to a large extent through control of protein stability. The MDM2 protein has been shown to play a key role in two ways: as a result of its physical interaction with p53, MDM2 both represses p53 transcriptional activity and mediates the degradation of p53 (Lohrum and Vousden, 1999). Exposure of the cell to DNA damage and other p53-activating signals results in phosphorylation and acetylation events that inhibit MDM2 mediated degradation of p53 and lead to its stabilization and activation. On the other hand, active p53 binds to the mdm2 gene to stimulate its transcription. This dual relationship establishes a potent autoregulatory loop that has been considered an important mechanism to control the duration and the strength of p53 mediated responses to cellular stressors. The identification of the p53 related proteins p73 and p63, the coexistence in the same cell of both transcriptionally active (TA-) and dominant negative (DN-) isoforms of both p73 and p63 (Pozniak et al., 2000; Grob et al., 2001; Ishimoto et al., 2002) and the observation that MDM2 actually stabilizes TA-p73 proteins, rather than inducing their degradation (Lohrum and Vousden, 1999), has made the scenario far more complex. The demonstration of a p53-dependent activation of the P2-p73 promoter leading to the expression and accumulation of DN-p73 proteins, that in turn selectively downregulate their own promoter as well as the p21 promoter, represents an additional mechanism to ensure the extinction of the p53/p73 dependent response to DNA damage. Importantly this regulatory loop is activated only under conditions of repairable DNA damage and cells are allowed to re-enter the cell cycle and it is already put into motion when the p53 stabilization has not yet been counteracted by either the increase of MDM2 or other mechanisms. Thus, our results identify a new paradigm in the regulation of the p53/p73 cellular response to DNA damage.

References

  1. Agami R, Blandino G, Oren M, Shaul Y . 1999 Nature 399: 809–813

  2. Caplen NJ, Parrish S, Imani F, Fire A, Morgan RA . 2001 Proc. Natl. Acad. Sci. USA 98: 9742–9747

  3. Costanzo A, Merlo P, Pediconi N, Fulco M, Sartorelli V, Cole PA, Fontemaggi G, Fanciulli M, Blandino G, Balsano C, Levrero M . 2002 Mol. Cell. 9: 175–186

  4. De Laurenzi V, Costanzo A, Barcaroli D, Terrinoni A, Falco M, Annicchiarico-Petruzzelli M, Levrero M, Melino G . 1998 J. Exp. Med. 188: 1763–1768

  5. De Laurenzi VD, Catani MV, Terrinoni A, Corazzari M, Melino G, Costanzo A, Levrero M, Knight RA . 1999 Cell Death Differ. 6: 389–390

  6. Elbashir SM, Harborth J, Lendeckel W, Yalcin A, Weber K, Tuschl T . 2001 Nature 411: 494–498

  7. Gong JG, Costanzo A, Yang HQ, Melino G, Kaelin Jr WG, Levrero M, Wang JY . 1999 Nature 399: 806–809

  8. Grob TJ, Novak U, Maisse C, Barcaroli D, Luthi AU, Pirnia F, Hugli B, Graber HU, De Laurenzi V, Fey MF, Melino G, Tobler A . 2001 Cell Death Differ. 8: 1213–1223

  9. Hartwell LH, Kastan MB . 1994 Science 266: 1821–1828

  10. Irwin M, Marin MC, Phillips AC, Seelan RS, Smith DI, Liu W, Flores ER, Tsai KY, Jacks T, Vousden KH, Kaelin Jr WG . 2000 Nature 407: 645–648

  11. Ishimoto O, Kawahara C, Enjo K, Obinata M, Nukiwa T, Ikawa S . 2002 Cancer Res. 62: 636–641

  12. Levrero M, De Laurenzi V, Costanzo A, Gong J, Melino G, Wang JY . 1999 Cell Death Differ. 6: 1146–1153

  13. Levrero M, De Laurenzi V, Costanzo A, Gong J, Wang JY, Melino G . 2000 J. Cell. Sci. 113: 1661–1670

  14. Liefer KM, Koster MI, Wang XJ, Yang A, McKeon F, Roop DR . 2000 Cancer Res. 60: 4016–4020

  15. Lipardi C, Wei Q, Paterson BM . 2001 Cell 107: 297–307

  16. Lissy NA, Davis PK, Irwin M, Kaelin WG, Dowdy SF . 2000 Nature 407: 642–645

  17. Lohrum MAE, Vousden KH . 1999 Cell Death Differ. 6: 1162–1168

  18. Nishikura K . 2001 Cell 107: 415–418

  19. Oda K, Arakawa H, Tanaka T, Matsuda K, Tanikawa C, Mori T, Nishimori H, Tamai K, Tokino T, Nakamura Y, Taya Y . 2000 Cell 102: 849–862

  20. Pozniak CD, Radinovic S, Yang A, McKeon F, Kaplan DR, Miller FD . 2000 Science 289: 304–306

  21. Prives C, Hall PA . 1999 J. Pathol. 187: 112–126

  22. Rich T, Allen RL, Wyllie AH . 2000 Nature 407: 777–783

  23. Sayan AE, Sayan BS, Findikli N, Ozturk M . 2001 Oncogene 20: 5111–5117

  24. Sijen T, Fleenor J, Simmer F, Thijssen KL, Parrish S, Timmons L, Plasterk RH, Fire A . 2001 Cell 107: 465–476

  25. Yang A, Kaghad M, Wang Y, Gillett E, Fleming MD, Dotsch V, Andrews NC, Caput D, McKeon F . 1998 Mol. Cell 2: 305–316

  26. Yuan ZM, Shioya H, Ishiko T, Sun X, Gu J, Huang YY, Lu H, Kharbanda S, Weichselbaum R, Kufe D . 1999 Nature 399: 814–817

  27. Zhou BB, Elledge SJ . 2000 Nature 408: 433–439

Download references

Acknowledgements

This work was supported by grants from AIRC, MURST-Cofin, Telethon Projects A1072 and E1325 and Schering-Plough to M Levrero. S Vossio and F Moretti are supported by fellowships from the Fondazione A. Cesalpino. A Costanzo is supported by a Staff Scientist Grant from the Fondazione A. Cesalpino.

Author information

Correspondence to Massimo Levrero or Antonio Costanzo.

Rights and permissions

Reprints and Permissions

About this article

Keywords

  • p53
  • p73
  • oncogenes
  • tumor suppressor
  • apoptosis
  • cell cycle

Further reading