Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

p53- and p21-dependent premature APC/C–Cdh1 activation in G2 is part of the long-term response to genotoxic stress


The long-term cellular response to DNA damage is controlled by the tumor suppressor p53. It results in cell-cycle arrest followed by DNA repair and, depending on the degree of damage inflicted, premature senescence or apoptotic cell death. Here we show that in normal diploid fibroblasts the ubiquitin ligase anaphase-promoting complex or cyclosome (APC/C)–Cdh1 becomes prematurely activated in G2 as part of the sustained long-term but not the rapid short-term response to genotoxic stress and results in the degradation of numerous APC/C substrates. Using HCT116 somatic knockout cells we show that mechanistically premature APC/C activation depends on p53 and its transcriptional target p21 that mediates the signal through downregulation of the APC/C inhibitor Emi1. Cdc14B is dispensable in this setting but might function redundantly. Our data suggest an unexpected role for the APC/C in executing a part of the p53-dependent DNA damage response that leads to premature senescence.


During their lifetime, cells are exposed to a variety of agents that cause damage to their DNA. An integral part of the cellular response to DNA damage is the ability to arrest DNA synthesis and cell-cycle progression. This arrest can be rapidly induced by executing a cascade of (de-)phosphorylation events that lead to an inactivation of cyclin-dependent kinases (CDKs) and replication factors. This immediate response is only transient and relieved when DNA damage has been repaired. However, in the case of more severe damage, the transient cell-cycle arrest gives way to either apoptotic cell death or a delayed but long-lasting, sometimes irreversible arrest. This long-term arrest can be established in G1 and G2 phase and relies on the tumor-suppressor protein-53 (p53)-dependent transcriptional activation of p21Waf1/Cip1 (p21) (Di Leonardo et al., 1994; Waldman et al., 1995; Bunz et al., 1998), a potent inhibitor of CDK activity (Harper et al., 1993; Besson et al., 2008). Cells arrested terminally after p21 induction acquire a senescence-like phenotype, also called premature senescence. Although DNA damage-induced apoptosis typically also depends on p53, p53 target genes other than p21 are involved (Yu and Zhang, 2005). Thus, p21 has an antiapoptotic function (Polyak et al., 1996; Janicke et al., 2007) and is actively repressed in apoptotic cells (Allan and Fried, 1999; Seoane et al., 2002). Both, DNA damage-induced apoptosis and senescence are considered to be important physiological mechanisms to withdraw pre-neoplastic cells from proliferation (Halazonetis et al., 2008). In addition, the outcome of cytotoxic cancer therapy crucially depends on the ability of tumor cells to elicit either an apoptotic or senescence response (Schmitt, 2003).

As a transcription factor, p53 has multiple regulatory functions. Besides its role as an activator of apoptosis and senescence-inducing genes, p53 also acts as a direct transcriptional repressor of several cell-cycle-promoting factors (St Clair et al., 2004; Banerjee et al., 2009). Moreover, through p21-dependent inhibition of CDK activity, p53 indirectly induces the formation of repressive pocket protein–E2F complexes, thereby leading to sustained transcriptional downregulation of many key enzymes of mitosis and replication (Löhr et al., 2003; Jackson et al., 2005; Spurgers et al., 2006; Kidokoro et al., 2008). This negative effect on mRNA synthesis is complemented at the level of protein production by another class of p53 target genes, the miR-34 family (He et al., 2007; Hermeking, 2007).

Given the broad influence of p53 on transcription and translation, it is surprising that at the level of protein degradation, a p53-driven mechanism of gene regulation is yet unknown. Such a mechanism might be particularly important for the long-term establishment of a stress-induced G2 arrest because in this cell-cycle phase cells have already accumulated significant amounts of mitotic enzymes. In principle, the anaphase-promoting complex or cyclosome (APC/C) would represent an attractive target for such a p53-dependent regulation, as this ubiquitin ligase functions as a master regulator for the stability of several rate-limiting proteins controlling replication and cell division (Peters, 2006).

Indeed, two reports have shown that in response to DNA damage the APC/C, which is normally kept inactive from the late G1 to early M-phase, is prematurely activated in G2 (Sudo et al., 2001; Bassermann et al., 2008). However, this activation followed DNA damage with short-term kinetics (as early as 90 min), making a p53-driven mechanism involving de novo protein synthesis unlikely. Moreover, except for polo-like kinase-1 (Plk1), both reports failed to observe a significant downregulation of APC/C substrates during the first 8 h after DNA damage, suggesting that APC/C activation has only a confined role during the early response to DNA damage.

Here we show that APC/C activation in G2-arrested cells is part of the p53- and p21-dependent long-term response to genotoxic stress that leads to premature senescence. It relies on Cdh1, the G1-phase-specific adaptor subunit (Wäsch et al., 2010), and goes hand in hand with marked decline of Emi1 expression, the main interphase inhibitor of the APC/C (Hsu et al., 2002). Importantly, activation of APC/C–Cdh1 is required for timely destruction of a broad range of APC/C substrates after DNA damage.


A senescence-like phenotype of doxorubicin-treated fibroblasts in G2

To investigate the cellular response to genotoxic stress at the level of APC/C regulation we used primary fibroblasts, as most transformed cell lines have acquired genetic alterations in DNA damage checkpoint pathways (Bartek et al., 2007). In particular, we wanted to address whether the silent APC/C of S/ G2 cells could be activated in response to DNA damage. Therefore, we synchronized cells in S-phase before applying DNA-damaging agents. To avoid checkpoint activation by chemical inhibitors of DNA replication (Nayak and Das, 2002; Kurose et al., 2006), synchronization in G1 was achieved by contact inhibition and replating of cells at lower density. Using this protocol, 17–19 h after replating 70–80% of cells entered S-phase synchronously (Figure 1a) that well suffices the requirements of this study.

Figure 1

Doxorubicin-induced long-term G2 arrest is accompanied by activation of the APC/C. Primary lung fibroblasts were first synchronized in G1 by density arrest and then re-plated at lower cell density to allow re-entry into the cell cycle. Nineteen hours after re-plating when most cells have reached mid-to-late S-phase (0 h), they were treated for 2 h with 1 μM doxorubicin or left untreated. Between 0 and 36 h after doxorubicin treatment, the cells were harvested at regular intervals and analyzed for cell-cycle position (a, b, d), DNA damage checkpoint activation (b), APC/C activity (c) and expression of APC/C substrates and regulators (b). (a) Cellular DNA content was analyzed by flow cytometry of propidium iodide-stained cells. Shown are DNA histograms where the relative cell number is plotted against DNA content (n=haploid number of chromosomes). (b) Whole-cell lysates were analyzed for the expression and/or the phosphorylation state of the indicated proteins by immunoblot analysis. (c) The APC/C was immunoprecipitated from cell lysates and tested for ubiquitin ligase activity using methylated ubiquitin (met-Ubi) and radioactively labeled cyclin-B1 as substrates. (d) Cell nuclei were analyzed by phase-contrast and fluorescence microscopy using 4′,6-diamidino-2-phenylindole (DAPI) stain and Alexa-488-labeled lamin-A/C-specific antibodies. APC/C, anaphase-promoting complex or cyclosome.

Moderate DNA damage of these cells was induced with 1 μM doxorubicin treatment for 2 h, causing a block to DNA replication. After removal of the drug, cells resumed cell-cycle progression but failed to undergo cell division and instead accumulated 4n-DNA content between 6 and 18 h after treatment (Figure 1a). Markers of mitotic entry as histone-H3–serine-10 phosphorylation (Figure 1b) and chromatin condensation (Figure 1d) were not detectable, showing that doxorubicin-treated cells were arrested in G2 rather than mitosis. We further ruled out a tetraploid G1 arrest as we detected no bi- or micro-nucleated cells (Figure 1d) that would origin from aberrant mitosis (Andreassen et al., 2001a, 2001b). The G2 arrest was remarkably stable lasting for at least 4 weeks (Supplementary Figure S1A). In addition, after 2–3 days the arrested cells developed features of cellular senescence such as β-galactosidase (SA-β-gal) activity and increase in both cell and nuclear size (Supplementary Figure S1B–D). This is consistent with doxorubicin-induced senescence in tumor cells (Chang et al., 1999) and primary fibroblasts (Baus et al., 2003; Demidenko and Blagosklonny, 2008).

Both, the transient arrest in S and the prolonged block in G2 were reflected by the expected biochemical changes to DNA damage checkpoint proteins. The early response was characterized by induction of H2A-X, Chk1, Chk2 and p53 phosphorylation, and of p53 abundance (Figure 1b). Six hours after doxorubicin treatment, these changes progressively decreased whereas expression of the long-term cell-cycle arrest effector p21 increased, reaching a steady-state level at around 24 h after DNA damage (Figure 1b and Supplementary Figure S3). Thus, the experimental system at hand faithfully allowed to differentiate between the immediate transient and a delayed sustained cell-cycle response to genotoxic stress.

Premature activation of the APC/C in doxorubicin-arrested G2 cells

We next examined whether the APC/C becomes activated under these conditions and if so, which kinetics would the activation follow. We did not observe significant APC/C activation early after DNA damage (Figure 1c), even when focusing on the first 90 min of doxorubicin treatment or DNA damage induction in G2 rather than in S-phase cells (Supplementary Figure S2). Instead, we found induction of APC/C activity that parallels the long-term DNA damage response beginning between 6 and 12 h and, like p21 protein expression, reaching a maximum at around 24 h after doxorubicin treatment (Figure 1c). Importantly, this APC/C activation appeared to be functionally relevant. All bona fide APC/C substrates analyzed were found to be downregulated with kinetics similar to the observed APC/C activation (Figure 1b). This included Cdc20, the M-phase-specific adaptor subunit of the APC/C, and other substrates of the APC/C–Cdh1, suggesting that premature APC/C activation in G2 was mediated by Cdh1, which remained stably expressed (Figure 1b) and co-immunoprecipitated with the APC/C (data not shown). Expression of UbcH10 was also affected by the DNA damage response, whereas another APC/C-interacting ubiquitin conjugase, UbcH5, was upregulated. Possibly, UbcH5, which compared with UbcH10 is considered as the less APC/C-specific E2 enzyme (Summers et al., 2008), has a role in the DNA damage-dependent induction of APC/C activity.

Premature induction of APC/C activity in G2 depends on Cdh1 and is required for timely destruction of APC/C substrates

Many APC/C substrate-encoding genes are known to underlie additional cell-cycle- and checkpoint-dependent regulatory mechanisms, that is, at the level of transcription and mRNA stability. Consistent with this notion, we found that after doxorubicin treatment APC/C targets were also downregulated at the mRNA level (Figure 2). To analyze the extent to which the observed downregulation of APC/C substrates could be attributed to an APC/C-dependent decrease in protein stability, we examined the DNA damage response in fibroblasts with stable and efficient Cdh1 knockdown (Figure 3c). In these cells, doxorubicin-dependent APC/C activation was completely prevented (Figure 3b), confirming our assumption that the APC/C becomes prematurely activated in its Cdh1-bound form. As shown for cyclin-A and B, knockdown of Cdh1 did not significantly influence the steady-state mRNA levels of the APC/C targets (Figure 3d) that finally become downregulated after DNA damage irrespectively of available APC/C activity. By contrast, lack of APC/C–Cdh1 activation led to a significant delay (6–12 h) in the DNA damage-dependent decrease of cyclin-A and B protein levels (Figure 3c). Thus, in line with the in vitro-measured increase in APC/C enzymatic activity, degradation of APC/C targets in vivo is of apparent importance to the efficient and particularly the prompt decline of cell-cycle-related gene products in doxorubicin-treated S/ G2 cells.

Figure 2

Emi1 and APC/C substrates are repressed at the mRNA level after DNA damage. RNA preparations from the experiment described in Figure 1 were analyzed by ribonuclease protection assay for mRNA expression of the indicated APC/C substrates (a, b) and of Emi1 (c). L32 and GAPDH were analyzed to control for equal loading. A radiolabeled probe was loaded to control for complete digestion of single-stranded RNA. In addition, a tRNA control served to detect nonspecific hybridization products as the one indicated by asterisk in panel c. APC/C, anaphase-promoting complex or cyclosome; GAPDH, glyceraldehyde-3-phosphate dehydrogenase.

Figure 3

The APC/C–Cdh1 is required for timely destruction of APC/C substrates in DNA damage-induced G2 arrest. Primary fibroblasts were stably transduced with lentiviral expression vectors for Cdh1- or GFP-specific short-hairpin RNAs. The resulting knockdown cells (Cdh1-KD, GFP-KD) and mock-infected control cells were treated with doxorubicin (doxo) as detailed in the legend to Figure 1. After doxorubicin treatment, the cells were harvested at 6-h intervals and analyzed for cell-cycle distribution (a), APC/C activity (b), protein (c) and mRNA expression (d) as described in the legends to Figures 1 and 2 respectively. APC/C, anaphase-promoting complex or cyclosome; GFP, green fluorescent protein.

DNA damage efficiently downregulates the APC/C inhibitor Emi1

The APC/C is normally kept inactive during S and G2 by binding to its pseudosubstrate inhibitor Emi1. In doxorubicin-treated S/ G2 cells Emi1 protein expression was found to be drastically reduced with kinetics inversely correlated to p21 expression (Figure 1b). This decline was detectable before an effect at the Emi1 mRNA levels (Figure 2c), suggesting that Emi1 is also targeted for protein degradation by a DNA damage-induced mechanism. Indeed, Emi1 protein expression in doxorubicin-treated cells was stabilized by proteasome inhibition (Supplementary Figure S4B). Surprisingly, this rescue of Emi1 expression did not prevent APC/C activation (Supplementary Figures S4D and S5A) although Emi1 became efficiently bound to the APC/C (Supplementary Figure S5B). Thus, Emi1 is downregulated after DNA damage and in addition mechanisms independent of Emi1 availability contribute to activate the APC/C after genotoxic stress.

Premature APC/C activation after γ-irradiation

To analyze whether premature APC/C activation is specific to chemical induction of DNA damage, we next used γ-ray doses between 2 and 100 Gy on S-phase cells and analyzed their short- and long-term response in comparison with doxorubicin treatment. Both agents had very similar effects. Again, the initial response was characterized by a reduction in DNA replication (Figure 4a), ongoing DNA repair and activation of checkpoint kinases as indicated by histone-H2A-X and Chk1 phosphorylation, respectively (Figure 4c). These effects had ceased by 24 h when a G2 arrest with high p21 expression (Figures 4a and c) and no visible signs of apoptosis (data not shown) was established. Importantly, at these time points the APC/C became activated also in γ-irradiated cells (Figure 4b). This activation was dose-dependent and accompanied by a decrease in Emi1 and APC/C–Cdh1 substrates as seen in doxorubicin-treated cells. Interestingly, activation of the APC/C and downregulation of Emi1 and APC/C substrates correlated even stronger with p21 rather than p53 expression, suggesting a possible link between p21 induction and APC/C regulation. Thus, activation of the APC/C in G2-arrested cells is a general long-term consequence of applied genotoxic stress and correlates well with p21 induction.

Figure 4

γ-Irradiation leads to premature APC/C activation in G2. Primary lung fibroblasts were synchronized in S-phase as described in Figure 1 and treated with either doxorubicin or γ-irradiation at the indicated doses. Cells were harvested at early (4 h) and late (24 h) time points after DNA damage and analyzed for cell-cycle distribution (a), APC/C activity (b) and the expression/phosphorylation state of the indicated proteins (c) as described in Figure 1. APC/C, anaphase-promoting complex or cyclosome.

Premature activation of the APC/C is p53-dependent

Given the strong correlation between the DNA damage-induced premature APC/C activation and p21 expression we reasoned that APC/C activation might mechanistically occur in a p53-dependent manner. Short-hairpin RNA-mediated silencing of p53 in primary fibroblasts abolished both p53 expression and, as would be expected, induction of p21 after doxorubicin treatment (Figure 5c). In agreement with the literature (Bunz et al., 1998), p53-deficient cells could still respond to DNA damage with G2 arrest (Figure 5a), possibly by keeping Cdk1 inactive through Wee1-dependent phosphorylation (Figure 5c). However, cells are impaired to maintain this arrest and underwent mitotic catastrophe after 2–3 days (Bunz et al., 1998; data not shown).

Figure 5

Premature induction of APC/C activity after DNA damage depends on the p53 tumor suppressor. Primary fibroblasts with a stable knockdown of p53 (p53-KD) and the indicated control fibroblasts (mock: mock-infected; GFP-KD: knockdown of green fluorescent protein) were synchronized and treated with doxorubicin (doxo) as described in Figure 1. Cells were harvested at 6, 12, 24, 48 and 72 h after doxorubicin treatment and analyzed for cell-cycle distribution (a), APC/C activity (b) and protein expression (c) as described in Figure 1. APC/C, anaphase-promoting complex or cyclosome; p53, tumor-suppressor protein-53.

p53-knockdown cells failed to activate the APC/C after doxorubicin treatment (Figure 5b) showing that APC/C induction after genotoxic stress depends indeed on p53 but is dispensable for installing the observed G2 arrest. Consistent with this observation, APC/C substrates, including cyclin-B1 and Aurora-A proteins, were not downregulated as in control cells but rather accumulated to high levels (Figure 5c). Interestingly, Emi1 downregulation was also prevented by the lack of p53, suggesting that both Emi1 protein and APC/C activity are targeted in a p53-dependent manner.

p21 but not Cdc14B is required for premature APC/C activation in HCT116 cells

To further mechanistically investigate the link between p53 and the APC/C, we used the human colon carcinoma cell line HCT116. As HCT116 cells are genetically stable and contain wild-type (wt) p53, they have been used to generate somatic knockouts of p53 and of genes functionally interacting with p53. Like p53-knockdown fibroblasts both HCT116-wt and p53-knockout (p53−/−) cells were arrested in G2 in response to doxorubicin treatment (Figure 6a) despite absence of p53 and p21 expression (Figure 6c). In wt cells establishment of the G2 arrest was also followed by induction of APC/C activity (Figure 6b), with the same late time kinetics as in primary fibroblasts. Premature APC/C activation was of a similar strength as the regular APC/C activation that occurred when untreated control cells divided and re-entered G1. By contrast, p53−/− cells were unable to activate the APC/C after DNA damage, although in untreated cells, G1-dependent APC/C activation was still preserved. After DNA damage Emi1 and APC/C substrates were repressed only in p53-positive cells (Figure 6c), whereas Emi1, Cdc20 and Aurora-A were reduced in G1 cells regardless of the p53 status. Thus, the main characteristics of DNA damage-dependent APC/C activation in G2 are conserved between primary lung fibroblasts and HCT116 cells, qualifying this transformed cell line as a suitable experimental system for further analysis.

Figure 6

p53-dependent APC/C activation in a transformed epithelial cell line. HCT116-wt cells and the isogenic p53-knockout strain (HCT116-p53−/−) were first synchronized in G1 by density arrest and then re-plated at lower cell density to allow reentry into the cell cycle. Fifteen hours after replating when most cells have reached mid-S-phase (0 h), they were treated with doxorubicin or left untreated. Between 0 and 36 h after doxorubicin treatment, cells were harvested at regular intervals and analyzed for cell-cycle distribution (a), APC/C activity (b) and protein expression (c) as described in Figure 1. A nonspecific protein band is indicated by asterisk. APC/C, anaphase-promoting complex or cyclosome; p53, tumor-suppressor protein-53; wt, wild type.

To directly investigate the question which p53-dependent pathway leads to APC/C activation in G2, we took advantage of HCT116 lines with isogenic knockouts of genes acting downstream (p21, 14-3-3σ) from p53 in the DNA damage checkpoint. All cell lines were able to respond to doxorubicin treatment with a G2 arrest albeit with varying efficiency (Figure 7a). Only wt and 14-3-3σ−/− cells showed induction of APC/C activity at 24 h after DNA damage (Figure 7b), implying that p21 is a crucial mediator of p53-dependent APC/C activation in G2. This is consistent with the above results showing close correlation between p21 induction and APC/C activity. Consequently, both p21 and p53 knockout had a strong negative effect on APC/C substrate degradation (Figure 7c). Similarly, doxorubicin-induced downregulation of Emi1 was completely abolished in both p53- and p21-negative cells. This further supports the view that Emi1 downregulation has a role in the p53-dependent induction of APC/C activity.

Figure 7

p21 but not Cdc14B is necessary for long-term APC/C activation after DNA damage. HCT116-wt cells and the indicated knockout/knockdown derivatives were synchronized and treated with doxorubicin as described in Figure 6. At 0, 6 and 24 h after doxorubicin treatment cells were harvested and analyzed for cell-cycle distribution (a), APC/C activity (b, e, g) and protein expression (c, h) as described in Figure 1. Cdc14B knockout and Cdc14B knockdown were confirmed by quantitative real-time PCR analysis of Cdc14B mRNA expression (d, f). APC/C, anaphase-promoting complex or cyclosome; wt, wild type.

Recently, a Cdc14B-knockout HCT116 derivative has been presented (Berdougo et al., 2008). The same factor was previously shown to be required for premature APC/C activation in doxorubicin-treated U2OS cells, by using an short interfering RNA-based knockdown approach (Bassermann et al., 2008). Surprisingly, Cdc14B−/− cells were still capable of activating the APC/C after doxorubicin treatment to the same extent as wt cells (Figure 7e). To rule out that the knockout clone we used had acquired compensatory mutations in the APC/C activatory pathway, we further performed a knockdown experiment using the same short interfering RNA that was used by Bassermann et al. and achieved reproducibly 80% silencing of Cdc14B at the mRNA level (Figure 7f), and this approach reproduced the results we achieved using the Cdc14B−/− cell line (Figure 7g). In summary, this set of data shows that APC/C activation after genotoxic stress strictly depends on p21 and that Cdc14B is not a limiting factor for premature APC/C activation in HCT116 cells.

Depleting Emi1 from p53−/− cells is sufficient to constitutively activate the APC/C in S/ G2

The inability of p53- and p21-deficient cells to activate the APC/C in response to genotoxic stress goes hand in hand with impaired downregulation of Emi1 (Figures 5,6,7). This led us to examine whether Emi depletion would restore the induction of APC/C activity in p53-knockout cells. Transfection of an Emi1-specific but not of a nonspecific (NS) short interfering RNA led to efficient silencing of Emi1 expression in HCT116-p53−/− cells (Figure 8c). Consistent with published data (Di Fiore and Pines, 2007; Machida and Dutta, 2007) the Emi1 knockdown predisposes cells to over-replication of their cellular genome (Figure 8a). The remainder of cells with a normal DNA content were still able to arrest in G2 after doxorubicin treatment (Figure 8a). Importantly, after Emi1 knockdown, G2-arrested, p53−/− cells showed the same level of APC/C activity as wt cells (Figure 8b). However, this was no de novo activation, as compared with S-phase cells at the beginning of doxorubicin treatment (0 h), APC/C activity was not further elevated. This shows that depletion of Emi1 in p53−/− cells resulted in constitutive APC/C activation that was independent of the cell-cycle position and persisted after DNA damage. Consistent with these data, the protein levels of APC/C substrates such as cyclin-B1 and Cdc20 were strongly reduced both in cycling and G2-arrested, Emi1-knockdown cells (Figure 8c). DNA over-replication, APC/C activation and destabilization of APC/C substrates were prevented by additional Cdh1 knockdown. These data emphasize the importance of APC/C–Cdh1 activity for the destruction of pro-mitotic factors during DNA damage-induced G2 arrest.

Figure 8

Emi1 depletion is sufficient to restore APC/C activation in p53-deficient cells. HCT116-wt and p53−/− cells were first transfected with the indicated short interfering RNAs and then, 3 days later, treated with doxorubicin as described in Figure 5. At 0 and 24 h after doxorubicin treatment cells were harvested and analyzed for cell-cycle distribution (a), APC/C activity (b) and protein expression as described in Figure 1 (c). The percentages of polyploid cells (DNA content >4n) are indicated (a). APC/C, anaphase-promoting complex or cyclosome; p53, tumor-suppressor protein-53; wt, wild type.


In this report we have studied the APC/C under conditions of an activated DNA damage checkpoint in the G2 phase of the cell cycle. We show that the APC/C ubiquitin ligase activity can be prematurely activated in G2 as part of a p53/p21-dependent long-term response to DNA damage. Both, in primary fibroblasts and in HCT116 cells, this activation translates into a broad downregulation of APC/C substrates, thereby providing an additional layer of p53-dependent gene repression during the cellular response to genotoxic stress. Our data are in good agreement with two recent studies that also used HCT116 cells to show p21-dependent destabilization of cyclin-B1 (Gillis et al., 2009; Lee et al., 2009), cyclin-A, Cdc20, Securin and Emi1 (Lee et al., 2009) after induction of DNA damage.

We did not, however, find experimental support for the rapid induction of APC/C activity that has been reported for X-irradiated HeLa cells (Sudo et al., 2001) and doxorubicin-treated U2OS cells (Bassermann et al., 2008). Possibly, the chemical inhibitors of DNA replication that were used in both studies of cell-cycle synchronization before induction of DNA damage particularly prime cells for rapid APC/C activation. In support of this notion, exposure to such compounds has been reported to elicit a DNA damage-like stress response (Nayak and Das, 2002; Kurose et al., 2006). Alternatively, the different APC/C substrates used in the in vitro ubiquitination assays, Cdc20 (Sudo et al., 2001), Plk1 (Bassermann et al., 2008) and cyclin-B1 (this study), might vary in their sensitivity to checkpoint-dependent changes in APC/C activity. With the exception of Plk1, both studies surprisingly failed to observe a significant decrease of APC/C substrates in vivo (Bassermann et al., 2008). Interestingly, we also found that, in contrast to primary fibroblasts and HCT116 cells, in G2-arrested U2OS cells most APC/C substrates resist APC/C-dependent degradation even at late time points after DNA damage (Supplementary Figure S6). Therefore, it appears that in U2OS and HeLa cells, cell-type-specific mechanisms might be in place that restrict APC/C activity. Possibly, deubiquitinases like USP28 (Zhang et al., 2006; Bassermann et al., 2008), lack of APC/C-specific E2 enzymes, weakened APC/C-proteasome interaction (Kob et al., 2009), a defect in the downregulation of APC/C substrate mRNAs and/or the moderate strength of premature APC/C induction (Supplementary Figure S6) could be responsible for the low effectiveness of APC/C activity in these cells.

Remarkably, we found that APC/C activation in G2 is associated with the induction of a senescence-like phenotype. Consistent with such a significant phenotypic change it is not surprising that the APC/C is activated in its Cdh1 and not the Cdc20-bound form. First, the APC/C–Cdh1 has a much broader substrate specificity, including many crucial activators of cell division and DNA replication that are not targeted by APC/C–Cdc20 (Peters, 2006; Wäsch et al., 2010). Second, because p21 is a substrate of the APC/C–Cdc20 in mitosis (Amador et al., 2007) and of SCFSkp2 during interphase (Yu et al., 1998; Bornstein et al., 2003; Wang et al., 2005), APC/C–Cdh1-dependent degradation of Cdc20 (Pfleger and Kirschner, 2000) and Skp2 (Bashir et al., 2004; Wei et al., 2004; Liu et al., 2007) might reinforce the p53-dependent induction of p21 expression after DNA damage. Given that p21 on the one hand leads to activation of APC/C–Cdh1 (this study) and on the other hand inhibits CDK-dependent phosphorylation, which otherwise would protect Skp2 from APC/C–Cdh1 targeting (Rodier et al., 2008), the existence of a positive feedback loop between p21 expression and APC/C–Cdh1 activation appears to be an attractive option and might be complemented and further enforced by p53/p21-dependent repression of Cdc20 transcription (Kidokoro et al., 2008).

An additional regulatory circuitry might be formed between p21, the APC/C and Cdc6. Like Skp2, Cdc6 is withdrawn from APC/C–Cdh1-dependent turnover in G1 by CDK2 phosphorylation (Mailand and Diffley, 2005). p53/p21-mediated inhibition of CDK activity after DNA damage abrogates this protection, thereby preventing Cdc6-dependent replication licensing (Duursma and Agami, 2005). The circuitry is completed by the finding that forced expression of Cdc6 can interfere with p21-dependent inhibition of CDK2 activity (Kan et al., 2008). Our finding of p21-dependent activation of APC/C–Cdh1 in G2 suggests that the same mechanisms controlling Cdc6 stability in G1 can be activated in G2-arrested cells and protect them from over-replication. In fact, p21 is an essential factor for the maintenance of normal DNA content in G2-arrested cells (Waldman et al., 1996). Tight control of Cdc6 expression in DNA-damaged cells might be particularly important in the absence of Emi1 expression that is known to trigger genome over-replication (Di Fiore and Pines, 2007; Machida and Dutta, 2007).

What links p21 to APC/C activation at the molecular level? So far Emi1 is the prime candidate as its downregulation during doxorubicin-induced G2 arrest does not only correlate with p21 induction but proved to be p53- and p21-dependent. In addition, we have shown that Emi1 silencing led to constitutive APC/C activation in p53-deficient cells. However, several lines of evidence also suggest that Emi1 is not the only factor that can mediate p21-dependent APC/C activation. First, prevention of Emi1 downregulation and APC/C–Emi1 dissociation by proteasome inhibition had no influence on APC/C activation (Supplementary Figures S4 and S5). This is reminiscent of Emi1 degradation not being required for APC/C activation in mitosis (Di Fiore and Pines, 2007). Second, in U2OS cells APC/C activation by DNA damage occurs despite inefficient downregulation of Emi1. Thus, in addition to Emi1 downregulation a functionally redundant mechanism must exist downstream from p21 to safeguard APC/C activation after DNA damage. Although Cdc14B is not required for APC/C activation in HCT116 cells (Figure 7), it remains an excellent candidate to work in parallel with Emi1. In that case Cdc14B regulation would not be required for APC/C activation when Emi1 downregulation is in place. Such a model (summarized in Supplementary Figure S7) is supported by the finding that Cdc14B is a necessary factor for premature APC/C activation in U2OS cells where Emi1 downregulation is not efficient (Bassermann et al., 2008).

It appears that p21-induced Emi1 reduction can be rationalized by at least two distinct mechanisms. Lee et al. (2009) provided evidence that the CDK-inhibitory function of p21 contributes to Emi1 repression through a retinoblastoma protein-dependent inhibition of Emi1 transcription. Consistent with this notion, we also observed downregulation of Emi1 mRNA levels in our system, but in addition found that disappearance of Emi1 protein precedes the decline of Emi1 mRNA. Also, proteasomal inhibition was able to prevent Emi1 downregulation. Thus, it seems that p21 interferes with Emi1 expression at two levels: increased degradation removes the pre-existing Emi1 protein and transcriptional silencing inhibits its de novo synthesis. Currently, we can only speculate as to how Emi1 is destabilized after DNA damage. During mitotic entry of the undisturbed cell cycle, Emi1 is marked for SCFβTrCP-dependent degradation by Plk1-dependent phosphorylation (Hansen et al., 2004; Moshe et al., 2004). Whereas Plk1 is inhibited by the DNA damage checkpoint (Smits et al., 2000) and can therefore not account for Emi1 destabilization, its homologues Plk2 and Plk3 are activated after genotoxic stress (Xie et al., 2001; Bahassi el et al., 2002; Matthew et al., 2007) and can phosphorylate Emi1 in vitro (L Wiebusch and C Hagemeier, unpublished observation). It is thus conceivable that after DNA damage Plk2 or Plk3 substitute for Plk1 and trigger βTrCP-dependent degradation of Emi1. Alternatively, one has to consider that CDK activity in S/ G2 has an as yet unrecognized positive influence on Emi1 protein stability that is compromised by the CDK-inhibitory function of p21. We are currently exploring the possible mechanisms of p21-dependent Emi1 degradation to close the gap in understanding how such important cell-cycle regulators as p21 and the APC/C are interconnected.

Materials and methods

See Supplementary information for complete methods.


  1. Allan LA, Fried M . (1999). p53-dependent apoptosis or growth arrest induced by different forms of radiation in U2OS cells: p21WAF1/CIP1 repression in UV induced apoptosis. Oncogene 18: 5403–5412.

    CAS  Article  Google Scholar 

  2. Amador V, Ge S, Santamaría PG, Guardavaccaro D, Pagano M . (2007). APC/C(Cdc20) controls the ubiquitin-mediated degradation of p21 in prometaphase. Mol Cell 27: 462–473.

    CAS  Article  Google Scholar 

  3. Andreassen PR, Lacroix FB, Lohez OD, Margolis RL . (2001b). Neither p21WAF1 nor 14-3-3sigma prevents G2 progression to mitotic catastrophe in human colon carcinoma cells after DNA damage, but p21WAF1 induces stable G1 arrest in resulting tetraploid cells. Cancer Res 61: 7660–7668.

    CAS  Google Scholar 

  4. Andreassen PR, Lohez OD, Lacroix FB, Margolis RL . (2001a). Tetraploid state induces p53-dependent arrest of nontransformed mammalian cells in G1 . Mol Biol Cell 12: 1315–1328.

    CAS  Article  Google Scholar 

  5. Bahassi el M, Conn CW, Myer DL, Hennigan RF, McGowan CH, Sanchez Y et al. (2002). Mammalian Polo-like kinase 3 (Plk3) is a multifunctional protein involved in stress response pathways. Oncogene 21: 6633–6640.

    Article  Google Scholar 

  6. Banerjee T, Nath S, Roychoudhury S . (2009). DNA damage induced p53 downregulates Cdc20 by direct binding to its promoter causing chromatin remodeling. Nucleic Acids Res 37: 2688–2698.

    CAS  Article  Google Scholar 

  7. Bartek J, Bartkova J, Lukas J . (2007). DNA damage signalling guards against activated oncogenes and tumour progression. Oncogene 26: 7773–7779.

    CAS  Article  Google Scholar 

  8. Bashir T, Dorrello NV, Amador V, Guardavaccaro D, Pagano M . (2004). Control of the SCF(Skp2-Cks1) ubiquitin ligase by the APC/C(Cdh1) ubiquitin ligase. Nature 428: 190–193.

    CAS  Article  Google Scholar 

  9. Bassermann F, Frescas D, Guardavaccaro D, Busino L, Peschiaroli A, Pagano M . (2008). The Cdc14B–Cdh1–Plk1 axis controls the G2 DNA-damage-response checkpoint. Cell 134: 256–267.

    CAS  Article  Google Scholar 

  10. Baus F, Gire V, Fisher D, Piette J, Dulic V . (2003). Permanent cell cycle exit in G2 phase after DNA damage in normal human fibroblasts. EMBO J 22: 3992–4002.

    CAS  Article  Google Scholar 

  11. Berdougo E, Nachury MV, Jackson PK, Jallepalli PV . (2008). The nucleolar phosphatase Cdc14B is dispensable for chromosome segregation and mitotic exit in human cells. Cell Cycle 7: 1184–1190.

    CAS  Article  Google Scholar 

  12. Besson A, Dowdy SF, Roberts JM . (2008). CDK inhibitors: cell cycle regulators and beyond. Dev Cell 14: 159–169.

    CAS  Article  Google Scholar 

  13. Bornstein G, Bloom J, Sitry-Shevah D, Nakayama K, Pagano M, Hershko A . (2003). Role of the SCFSkp2 ubiquitin ligase in the degradation of p21Cip1 in S phase. J Biol Chem 278: 25752–25757.

    CAS  Article  Google Scholar 

  14. Bunz F, Dutriaux A, Lengauer C, Waldman T, Zhou S, Brown JP et al. (1998). Requirement for p53 and p21 to sustain G2 arrest after DNA damage. Science 282: 1497–1501.

    CAS  Article  Google Scholar 

  15. Chang BD, Broude EV, Dokmanovic M, Zhu H, Ruth A, Xuan Y et al. (1999). A senescence-like phenotype distinguishes tumor cells that undergo terminal proliferation arrest after exposure to anticancer agents. Cancer Res 59: 3761–3767.

    CAS  Google Scholar 

  16. Demidenko ZN, Blagosklonny MV . (2008). Growth stimulation leads to cellular senescence when the cell cycle is blocked. Cell Cycle 7: 3355–3361.

    CAS  Article  Google Scholar 

  17. Di Fiore B, Pines J . (2007). Emi1 is needed to couple DNA replication with mitosis but does not regulate activation of the mitotic APC/C. J Cell Biol 177: 425–437.

    CAS  Article  Google Scholar 

  18. Di Leonardo A, Linke SP, Clarkin K, Wahl GM . (1994). DNA damage triggers a prolonged p53-dependent G1 arrest and long-term induction of Cip1 in normal human fibroblasts. Genes Dev 8: 2540–2551.

    CAS  Article  Google Scholar 

  19. Duursma A, Agami R . (2005). p53-dependent regulation of Cdc6 protein stability controls cellular proliferation. Mol Cell Biol 25: 6937–6947.

    CAS  Article  Google Scholar 

  20. Gillis LD, Leidal AM, Hill R, Lee PW . (2009). p21Cip1/WAF1 mediates cyclin B1 degradation in response to DNA damage. Cell Cycle 8: 253–256.

    CAS  Article  Google Scholar 

  21. Halazonetis TD, Gorgoulis VG, Bartek J . (2008). An oncogene-induced DNA damage model for cancer development. Science 319: 1352–1355.

    CAS  Article  Google Scholar 

  22. Hansen DV, Loktev AV, Ban KH, Jackson PK . (2004). Plk1 regulates activation of the anaphase promoting complex by phosphorylating and triggering SCFbetaTrCP-dependent destruction of the APC inhibitor Emi1. Mol Biol Cell 15: 5623–5634.

    CAS  Article  Google Scholar 

  23. Harper JW, Adami GR, Wei N, Keyomarsi K, Elledge SJ . (1993). The p21 Cdk-interacting protein Cip1 is a potent inhibitor of G1 cyclin-dependent kinases. Cell 75: 805–816.

    CAS  Article  Google Scholar 

  24. He L, He X, Lowe SW, Hannon GJ . (2007). MicroRNAs join the p53 network—another piece in the tumour-suppression puzzle. Nat Rev Cancer 7: 819–822.

    CAS  Article  Google Scholar 

  25. Hermeking H . (2007). p53 enters the microRNA world. Cancer Cell 12: 414–418.

    CAS  Article  Google Scholar 

  26. Hsu JY, Reimann JD, Sorensen CS, Lukas J, Jackson PK . (2002). E2F-dependent accumulation of hEmi1 regulates S phase entry by inhibiting APC(Cdh1). Nat Cell Biol 4: 358–366.

    CAS  Article  Google Scholar 

  27. Jackson MW, Aquarwal MK, Yang J, Bruss P, Uchiumi T, Aquarwal ML et al. (2005). p130/7p107/p105Rb-dependent transcriptional repression during DNA-damage-induced cell-cycle exit at G2 . J Cell Sci 118: 1821–1832.

    CAS  Article  Google Scholar 

  28. Janicke RU, Sohn D, Essmann F, Schulze-Osthoff K . (2007). The multiple battles fought by antiapoptotic p21. Cell Cycle 6: 407–413.

    Article  Google Scholar 

  29. Kan Q, Jinno S, Yamamoto H, Kobayashi K, Okayama H . (2008). ATP-dependent activation of p21WAF1/CIP1-associated Cdk2 by Cdc6. Proc Natl Acad Sci USA 105: 4757–4762.

    CAS  Article  Google Scholar 

  30. Kidokoro T, Tanikawa C, Furukawa Y, Katagiri T, Nakamura Y, Matsuda K . (2008). CDC20, a potential cancer therapeutic target, is negatively regulated by p53. Oncogene 27: 1562–1571.

    CAS  Article  Google Scholar 

  31. Kob R, Kelm J, Posorski N, Baniahmad A, von Eggeling F, Melle C . (2009). Regulation of the anaphase-promoting complex by the COP9 signalosome. Cell Cycle 8: 2041–2049.

    CAS  Article  Google Scholar 

  32. Kurose A, Tanaka T, Huang X, Traganos F, Darzynkiewicz Z . (2006). Synchronization in the cell cycle by inhibitors of DNA replication induces histone H2AX phosphorylation: an indication of DNA damage. Cell Prolif 39: 231–240.

    CAS  Article  Google Scholar 

  33. Lee J, Kim JA, Barbier V, Fotedar A, Fotedar R . (2009). DNA damage triggers p21WAF1-dependent Emi1 downregulation that maintains G2 arrest. Mol Biol Cell 20: 1891–1902.

    CAS  Article  Google Scholar 

  34. Liu W, Wu G, Li W, Lobur D, Wan Y . (2007). Cdh1-anaphase-promoting complex targets Skp2 for destruction in transforming growth factor beta-induced growth inhibition. Mol Cell Biol 27: 2967–2979.

    CAS  Article  Google Scholar 

  35. Löhr K, Möritz C, Contente A, Dobbelstein M . (2003). p21/CDKN1A mediates negative regulation of transcription by p53. J Biol Chem 278: 32507–32516.

    Article  Google Scholar 

  36. Machida YJ, Dutta A . (2007). The APC/C inhibitor, Emi1, is essential for prevention of rereplication. Genes Dev 21: 184–194.

    CAS  Article  Google Scholar 

  37. Mailand N, Diffley JF . (2005). CDKs promote DNA replication origin licensing in human cells by protecting Cdc6 from APC/C-dependent proteolysis. Cell 122: 915–926.

    CAS  Article  Google Scholar 

  38. Matthew EM, Yen TJ, Dicker DT, Dorsey JF, Yang W, Navaraj A et al. (2007). Replication stress, defective S-phase checkpoint and increased death in Plk2-deficient human cancer cells. Cell Cycle 6: 2571–2578.

    CAS  Article  Google Scholar 

  39. Moshe Y, Boulaire J, Pagano M, Hershko A . (2004). Role of Polo-like kinase in the degradation of early mitotic inhibitor 1, a regulator of the anaphase promoting complex/cyclosome. Proc Natl Acad Sci USA 101: 7937–7942.

    CAS  Article  Google Scholar 

  40. Nayak BK, Das GM . (2002). Stabilization of p53 and transactivation of its target genes in response to replication blockade. Oncogene 21: 7226–7229.

    CAS  Article  Google Scholar 

  41. Peters JM . (2006). The anaphase promoting complex/cyclosome: a machine designed to destroy. Nat Rev Mol Cell Biol 7: 644–656.

    CAS  Article  Google Scholar 

  42. Pfleger CM, Kirschner MW . (2000). The KEN box: an APC recognition signal distinct from the D box targeted by Cdh1. Genes Dev 14: 655–665.

    CAS  Google Scholar 

  43. Polyak K, Waldman T, He TC, Kinzler KW, Vogelstein B . (1996). Genetic determinants of p53-induced apoptosis and growth arrest. Genes Dev 10: 1945–1952.

    CAS  Article  Google Scholar 

  44. Rodier G, Coulombe P, Tanguay PL, Boutonnet C, Meloche S . (2008). Phosphorylation of Skp2 regulated by CDK2 and Cdc14B protects it from degradation by APC(Cdh1) in G1 phase. EMBO J 27: 679–691.

    CAS  Article  Google Scholar 

  45. Schmitt CA . (2003). Senescence, apoptosis and therapy—cutting the lifelines of cancer. Nat Rev Cancer 3: 286–295.

    CAS  Article  Google Scholar 

  46. Seoane J, Le HV, Massague J . (2002). Myc suppression of the p21(Cip1) Cdk inhibitor influences the outcome of the p53 response to DNA damage. Nature 419: 729–734.

    CAS  Article  Google Scholar 

  47. Smits VA, Klompmaker R, Arnaud L, Rijksen G, Nigg EA, Medema RH . (2000). Polo-like kinase-1 is a target of the DNA damage checkpoint. Nat Cell Biol 2: 672–676.

    CAS  Article  Google Scholar 

  48. Spurgers KB, Gold DL, Coombes KR, Bohnenstiehl NL, Mullins B, Meyn RE et al. (2006). Identification of cell cycle regulatory genes as principal targets of p53-mediated transcriptional repression. J Biol Chem 281: 25134–25142.

    CAS  Article  Google Scholar 

  49. St Clair S, Giono L, Varmeh-Ziaie S, Resnick-Silverman L, Liu WJ, Padi A et al. (2004). DNA damage-induced downregulation of Cdc25C is mediated by p53 via two independent mechanisms: one involves direct binding to the cdc25C promoter. Mol Cell 16: 725–736.

    Article  Google Scholar 

  50. Sudo T, Ota Y, Kotani S, Nakao M, Takami Y, Takeda S et al. (2001). Activation of Cdh1-dependent APC is required for G1 cell cycle arrest and DNA damage-induced G2 checkpoint in vertebrate cells. EMBO J 20: 6499–6508.

    CAS  Article  Google Scholar 

  51. Summers MK, Pan B, Mukhyala K, Jackson PK . (2008). The unique N terminus of the UbcH10 E2 enzyme controls the threshold for APC activation and enhances checkpoint regulation of the APC. Mol Cell 31: 544–556.

    CAS  Article  Google Scholar 

  52. Waldman T, Kinzler KW, Vogelstein B . (1995). p21 is necessary for the p53-mediated G1 arrest in human cancer cells. Cancer Res 55: 5187–5190.

    CAS  Google Scholar 

  53. Waldman T, Lengauer C, Kinzler KW, Vogelstein B . (1996). Uncoupling of S phase and mitosis induced by anticancer agents in cells lacking p21. Nature 381: 713–716.

    CAS  Article  Google Scholar 

  54. Wang W, Nacusi L, Sheaff RJ, Liu X . (2005). Ubiquitination of p21Cip1/WAF1 by SCFSkp2: substrate requirement and ubiquitination site selection. Biochemistry 44: 14553–14564.

    CAS  Article  Google Scholar 

  55. Wäsch R, Robbins JA, Cross FR . (2010). The emerging role of APC/CCdh1 in controlling differentiation, genomic stability and tumor suppression. Oncogene 29: 1–10.

    Article  Google Scholar 

  56. Wei W, Ayad NG, Wan Y, Zhang GJ, Kirschner MW, Kaelin Jr WG . (2004). Degradation of the SCF component Skp2 in cell-cycle phase G1 by the anaphase-promoting complex. Nature 428: 194–198.

    CAS  Article  Google Scholar 

  57. Xie S, Wu H, Wang Q, Cogswell JP, Husain I, Conn C et al. (2001). Plk3 functionally links DNA damage to cell cycle arrest and apoptosis at least in part via the p53 pathway. J Biol Chem 276: 43305–43312.

    CAS  Article  Google Scholar 

  58. Yu J, Zhang L . (2005). The transcriptional targets of p53 in apoptosis control. Biochem Biophys Res Commun 331: 851–858.

    CAS  Article  Google Scholar 

  59. Yu ZK, Gervais JL, Zhang H . (1998). Human CUL-1 associates with the SKP1/SKP2 complex and regulates p21(CIP1/WAF1) and cyclin D proteins. Proc Natl Acad Sci USA 95: 11324–11329.

    CAS  Article  Google Scholar 

  60. Zhang D, Zaugg K, Mak TW, Elledge SJ . (2006). A role for the deubiquitinating enzyme USP28 in control of the DNA-damage response. Cell 126: 529–542.

    CAS  Article  Google Scholar 

Download references


We thank Eli Berdougo, Philipp Haemmati, Christoph Hanski, Hans-Jürgen Heidebrecht, Prasad Jallepalli, Bert Vogelstein and Ralph Wäsch for the generous supply of reagents. We acknowledge Matthias Truss for help with lentivirus production and Ralf Uecker for excellent technical assistance. This work was supported by Grants HA1575/2-1 and WI2043/2-2 from the Deutsche Forschungsgemeinschaft (DFG).

Author information



Corresponding author

Correspondence to L Wiebusch.

Ethics declarations

Competing interests

The authors declare no conflict of interest.

Additional information

Supplementary Information accompanies the paper on the Oncogene website

Supplementary information

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Wiebusch, L., Hagemeier, C. p53- and p21-dependent premature APC/C–Cdh1 activation in G2 is part of the long-term response to genotoxic stress. Oncogene 29, 3477–3489 (2010).

Download citation


  • APC/C
  • p53
  • p21
  • Emi1
  • senescence
  • apoptosis

Further reading


Quick links