The p400/Tip60 ratio is critical for colorectal cancer cell proliferation through DNA damage response pathways

Article metrics

Abstract

The Tip60 histone acetyltransferase belongs to a multimolecular complex that contains many chromatin remodeling enzymes including the ATPase p400, a protein involved in nucleosomal incorporation of specific histone variants and that can directly or indirectly repress some Tip60-dependent pathways. Tip60 activity is critical for the cellular response to DNA damage and is affected during cancer progression. Here, we found that the ratio between Tip60 and p400 mRNAs is affected in most colorectal carcinoma. Strikingly, reversing the p400/Tip60 imbalance by Tip60 overexpression or the use of siRNAs resulted in increased apoptosis and decreased proliferation of colon-cancer-derived cells, suggesting that this ratio defect is important for cancer progression. Furthermore, we demonstrate that the p400/Tip60 ratio controls the oncogene-induced DNA damage response, a known anticancer barrier. Finally, we found that it is also critical for the response to 5-fluorouracil, a first-line treatment against colon cancer. Together, our data indicate that the p400/Tip60 ratio is critical for colon cancer cells proliferation and response to therapeutic drugs through the control of stress–response pathways.

Introduction

Chromatin structure affects all processes requiring access to the DNA double helix, including transcription and DNA repair. It can be regulated by various means, including the post-translational modifications of nucleosomal histones. Among these, histone acetylation has been extensively studied, because it is known to correlate with transcriptional activation. It is now clear that it is also important for DNA replication and repair. Not surprisingly, the cellular machineries that regulate histone acetylation are thus involved in the proper control of cell fate, and deregulation of some of these machineries has been described in human cancer (Miremadi et al., 2007).

The Tip60 histone acetyltransferase has been recently shown to be underexpressed in many human cancers from various origins (Lleonart et al., 2006; Gorrini et al., 2007). Moreover, in a model of tumor induction in mice, it has been shown to function as a haploinsufficient tumor suppressor, providing a causal link between Tip60 underexpression and tumorigenesis (Gorrini et al., 2007).

Tip60 belongs to a multimolecular complex, the so-called ‘Tip60 complex’, which is composed of up to 16 subunits (Ikura et al., 2000). This complex contains essential cofactors for Tip60 HAT activity, because the Tip60 complex can acetylate nucleosomes, whereas purified Tip60 cannot. This Tip60 complex contains other enzymes, such as the helicases Tip49a and Tip49b (Ikura et al., 2000) and the ATPase p400 (Fuchs et al., 2001), a protein recently shown to mediate incorporation of histone H2A.Z variant in mammals (Gevry et al., 2007). It is important to note that Tip60 can also acetylate non-histone proteins, including transcription factors, such as p53 (Sykes et al., 2006; Tang et al., 2006) or the DNA damage signaling protein ATM (Sun et al., 2005).

Tip60 was shown to be important for transcriptional control. In agreement with its function in histone acetylation, Tip60 is mainly involved in transcriptional activation of specific genes (Gaughan et al., 2001; Berns et al., 2004; Legube et al., 2004). Transcriptional regulation of some genes, such as E2F or c-myc-dependent genes, most likely involves the whole Tip60 complex (Frank et al., 2003; Taubert et al., 2004). However, it was recently shown that, on p21 promoter at least, Tip60 and p400 have antagonistic functions (Tyteca et al., 2006), with p400 functioning directly or indirectly as a repressor of Tip60 (meaning that p400 knockdown induces p21 expression in a Tip60-dependent way).

p400 has been cloned as a factor associated with the E1A viral protein and important for its transforming abilities (Fuchs et al., 2001). Accordingly, p400 function is involved in proliferation control: p400 is important for E1A- or UV-irradiation-induced apoptosis (Chan et al., 2005; Tyteca et al., 2006) and p400 knockdown induces senescence in untransformed human fibroblasts (Chan et al., 2005), processes which all could be related to the ability of p400 to suppress p21 expression (Chan et al., 2005; Tyteca et al., 2006).

Because Tip60 function is important for cancer progression, we analysed the expression of both Tip60 and its upstream regulator p400 in human cancer. We confirmed that Tip60 and p400 mRNAs are underexpressed in a coordinated manner in colon carcinoma compared to adjacent normal tissue, inducing a p400/Tip60 imbalance disfavoring Tip60 function. We show that restoring the normal p400/Tip60 ratio is sufficient to promote apoptosis and growth arrest of colon cancer cells as well as the cellular response to 5-fluorouracil (5-FU), a first-line treatment against colon carcinoma. All these data suggest that the p400/Tip60 ratio is critical for colon carcinoma cells proliferation and could be a valuable therapeutic target.

Results

The p400/Tip60 ratio is altered in most colon carcinoma

Given the importance of Tip60 in cancer, we analysed the expression of Tip60 and its associated protein p400 in a collection of colorectal carcinoma (see Supplementary data 1 for patient selection, quantification of gene expression and statistical analysis). By real-time PCR analysis, we found that Tip60 mRNA is underexpressed in 72 out of 74 cancers (T) compared to adjacent normal (N) tissue (Figure 1a; see Supplementary data 2 for all T/N ratios). This decrease is highly significant (P<0.000001, one-sided binomial exact test) and was expected because similar results were already reported (Lleonart et al., 2006; Gorrini et al., 2007). Importantly, sequencing of the entire Tip60 cDNA in four tumoral samples shows that Tip60, although underexpressed, is not mutated in tumors. Thus, our data extend previous findings indicating that Tip60 is underexpressed in colorectal cancer. Tip60 underexpression is observed irrespective of the tumoral stage (stage 1, P=0.000004; stage 2, P=0.000008; stage 3, P=0.000275 and stage 4, P=0.000040), suggesting that the decrease of Tip60 expression modification occurs early in tumorigenesis. We also found that p400 mRNA is significantly underexpressed in colorectal cancer (P=0.013258, one-sided binomial exact test) (Figure 1a). In 14 out of 74 tumors, p400 expression is decreased compared to adjacent normal tissue by twofold or more. We do not understand, for the moment, either the mechanism of this underexpression or the growth advantage, if any, given by this lower expression.

Figure 1
figure1

Coordinated deregulation of Tip60 and p400 mRNA expression in colorectal carcinoma. (a) Total RNA from colon carcinoma (T) or adjacent normal tissue (N) was prepared, reverse transcribed and analysed by real-time PCR for the relative expression of p400 (upper graph) and Tip60 (lower graph). T/N ratios of normalized values were calculated. T/N values lower than 1 were transformed into the inverse N/T values. Tumors were classified from the ones that overexpress the most p400 (left) to the ones that underexpress the most p400 (right). The representation of the relative expression of Tip60 in the corresponding tumors is shown in the lower graph. (b) The relative Tip60 expression in a given tumor is plotted against the relative p400 expression in the same tumor (both calculated as the base 2 logarithm of the T/N ratio). The graph and the equation of the linear regression are shown. (c) Nuclear extracts (60 μg) from the indicated cell lines were analysed by western blot for Tip60 or p400 expression. Also shown is a histone deacetylase (HDAC) western blot as a loading control (please note that HDAC1/2 expression is higher in cancer cell lines, in agreement with previous findings that HDAC1/2 mRNA is overexpressed in colon carcinoma (Wilson et al., 2006)).

Strikingly, when analysing Tip60 mRNA expression in the same tumors (Figure 1a), a strong correlation with p400 mRNA expression is observed (nonparametric Spearman's ρ correlation, ρ=0.8286, P<0.0001). Within the tumors collection, when the T/N ratio of Tip60 expression decreases by twofold (1 unit in Figure 1a), there is a very similar decrease of the T/N ratio of p400 expression (0.85 unit, which corresponds to 1.8-fold; Figure 1b). This result suggests that the ratio between p400 and Tip60 mRNA changes is more or less constant within the tumors collection. Moreover, it is increased by about twofold compared to normal tissue (Figure 1b). Because p400 can function as a repressor of some Tip60-dependent pathways, this change in the p400/Tip60 ratio could disfavor these pathways. These data are consistent with the possibility that a given level of inhibition of these Tip60-dependent pathways is achieved in most colon cancers: in the tumors such as the ones at the left of Figure 1a, this inhibition could be mediated by p400 overexpression. In tumors in the middle of the panel, p400 expression is not affected and the inhibition of Tip60-dependent pathways would be achieved by a mild decrease of Tip60 expression (approximately twofold). Tumors in which p400 is decreased (at the right of the panel) would achieve the inhibition of Tip60-dependent pathways by strongly decreasing Tip60 expression.

Thus, these data suggest that a change in the ratio between p400 and Tip60 mRNA expression (corresponding to a slight inhibition of Tip60-dependent pathways) is observed in most colon carcinomas. Because Tip60 functions as a haploinsufficient tumor suppressor, such changes in the p400/Tip60 ratio can thus be critical for colon cancers progression.

Tip60 function is rate limiting for apoptosis in HCT116 cells

We thus intended to modify this ratio in in vitro models of colon carcinoma. First, we checked several colorectal-cancer-derived cell lines for Tip60 and p400 expression by western blot. We found that in these cells also the ratio between p400 and Tip60 mRNA expression is increased compared to normal human intestinal epithelial cells (HIECs; Supplementary data 3). Moreover, we confirmed this finding for p400 and Tip60 proteins expression by western blot (Figure 1c) comparatively to non-colorectal cancer cells (HeLa) or to the normal HIECs (Figure 1c). Importantly, sequencing of the entire Tip60 mRNA in HCT116, HCT8 and HT29 cells indicates that it is not mutated. Thus, these cell lines likely experienced a change in the p400/Tip60 ratio and thus represent adequate models for investigating the consequence of this change.

We chose to focus on HCT116 cells, because they express Tip60 at about the same level as that of primary cancers (data not shown). We observed that Tip60 overexpression in HCT116 cells is detrimental to cell growth, because fewer colonies are observed in cells transfected by the Tip60-expressing vector compared to empty vector (Figure 2a). Interestingly, in a similar experiment performed in osteosarcoma U2-OS cells, Tip60 overexpression does not lead to growth suppression, probably because Tip60 function is not rate limiting for growth suppression in these cells. To investigate the mechanism responsible for this growth defect, we raised a pool of HCT116 cells stably overexpressing Tip60 (Supplementary data 4). In agreement with Figure 2a results, these cells grew more slowly than controls (data not shown). To discriminate between proapoptotic or cell-cycle-inhibiting effect of Tip60 overexpression, we analysed, by RT–QPCR, the expression of two proapoptotic genes (Fas and Bax) and the cell-cycle inhibitor p21, which are known targets of Tip60 (Figure 2b). Because Fas and Bax expression levels are stimulated by about 3.5- and 2.0-fold, respectively, whereas p21 remains unchanged, we postulated that Tip60 overexpression principally impacts apoptosis process. This was confirmed by poly-(ADP-ribose)polymerase (PARP) cleavage study and fluorescence-activated cell sorting (FACS) analysis of Annexin V binding (Figure 2c). Therefore, these data indicate that overexpression of Tip60 in HCT116 cells leads to increased apoptosis and growth defect, suggesting that Tip60 function in colorectal cancer is rate limiting for apoptosis.

Figure 2
figure2

Overexpression of Tip60 decreased HCT116 proliferation. (a) HCT116 cells (left) or U2-OS cells (right) were transfected with pcDNA3-HA-Tip60 or empty vector as a control, and subjected to a growth suppression assay. Cells were stained with crystal violet, and the numbers of clones were assessed. The mean and SDOM from three independent experiments are shown (relative to 1 for empty vector). (b) Total mRNA from an HCT116 cell population stably expressing Tip60 (white bars) and a control cell population (black bars) were extracted and subjected to reverse transcription followed by a QPCR analysis for Fas, Bax, p21 and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) mRNA expression. The amounts of Fas, Bax and p21 mRNA were then divided by the amount of control GAPDH mRNA and calculated relative to 1 for control cells. The mean and standard deviation from three independent experiments are shown. (c) Indicated HCT116 cell populations were tested for apoptosis induction by western blotting monitoring poly-(ADP-ribose)polymerase (PARP) cleavage (left) or fluorescence-activated cell sorting (FACS) analysis of Annexin V/7-AAD staining (right). The mean and SDOM from three independent experiments are shown (relative to 1 for apoptosis levels in Tip60-expressing cells).

The p400/Tip60 ratio controls apoptosis in HCT116 cells

To analyse directly the importance of the p400/Tip60 ratio in HCT116 cells proliferation, we intended to inhibit p400 expression and test whether we can favor endogenous Tip60 function and consequently apoptosis. We first tested whether activation of endogenous Tip60 is sufficient to promote apoptosis. We treated HCT116 cells with a histone deacetylase (HDAC) inhibitor (trichostatin A, TSA) to activate artificially most acetylation-dependent pathways (including the Tip60-dependent ones). As in many other cancer cells, TSA treatment strongly induces apoptosis (Figure 3a). Importantly, this apoptosis is inhibited by siRNA directed against Tip60 (described in Legube et al. (2004) and Tyteca et al. (2006)), indicating that Tip60 expression is, at least in part, required for TSA-induced apoptosis (the efficiency of the knockdown was checked by western blot and RT-QPCR (Figure 3b). This result thus shows that the Tip60-dependent apoptotic pathways are functional in these cells and that activating endogenous Tip60 function is sufficient to promote apoptosis.

Figure 3
figure3

p400 controls Tip60-dependent apoptosis in HCT116 cells. (a) HCT116 cells were transfected with the indicated siRNAs. After 36 h, trichostatin A (TSA, 400 ng/ml) was added or not. After 24 h, apoptosis was assessed by western blotting monitoring poly-(ADP-ribose)polymerase (PARP) cleavage (left) or Annexin V/7-AAD staining (right). The mean and standard deviation from three independent experiments are shown (relative to 1 in TSA-treated cells transfected with control siRNA). (b) HCT116 cells were transfected with the indicated siRNAs alone or in combination. After 48 h, nuclear (for Tip60 and HDAC1/2) or total (for p400 and glyceraldehyde-3-phosphate dehydrogenase (GAPDH)) protein extracts were prepared and subjected to western blot analysis to detect Tip60 and HDAC1/2 or p400 and GAPDH. On the Tip60 blot, the asterisk indicates a nonspecific band. Note that the mild increase in p400 levels in cells transfected by the Tip60 siRNA is not reproducible. Total mRNA from the same samples was also prepared, reverse transcribed and subjected to QPCR analysis for Tip60 or p400 mRNA expression, relative to GAPDH mRNA used as a control. Results were calculated relative to 1 for cells transfected with control siRNA. The mean and standard deviation from three independent experiments are shown. (c) HCT116 cells were transfected with the indicated siRNAs alone or in combination. After 72 h, apoptosis was assessed by Annexin V/7-AAD staining. Results were calculated relative to 1 for cells transfected with p400 siRNA. The mean and standard deviation from three independent experiments are shown.

Transfection of HCT116 cells with a p400 siRNA (already described in Tyteca et al. (2006)) decreases p400 mRNA levels by approximately twofold and significantly inhibits p400 protein expression without affecting Tip60 expression (Figure 3b). Strikingly, this siRNA is able to induce apoptosis in these cells (Figure 3c; P<0.043577) in agreement with the possibility that it favors Tip60-dependent pathways. Importantly, co-transfecting a Tip60 siRNA (which also decreases Tip60 mRNA levels by about twofold without having any significant impact on p400 expression (Figure 3b) and thus brings back the p400/Tip60 ratio to its level in untransfected HCT116 cells) abolishes the effects of p400 depletion (P<0.049334) without significantly changing the efficiency of p400 depletion (Figure 3b). Thus, taken together, these results indicate that the defective p400/Tip60 ratio observed in HCT116 cells leads a protection against apoptosis due to the inhibition of Tip60-dependent pathways.

The p400/Tip60 ratio controls the DNA damage response induced by oncogenic stress

We next investigated the mechanism by which this defective p400/Tip60 ratio can protect cells against apoptosis. A twofold decrease in Tip60 function can be detrimental to the DNA damage response (DDR) pathway induced by oncogenic stress (Gorrini et al., 2007). We thus investigated the efficiency of this pathway in HCT116 cells. We transfected HCT116 cells with an expression vector encoding for a hemagglutinin (HA)-tagged fusion protein between a modified ligand-binding domain of estrogen receptor (ER) and the E2F1 oncoprotein. This fusion protein translocates to the nucleus on hydroxytamoxifen (OH-Tam) addition (Supplementary data 5A), resulting in the induction of a DDR pathway in U2-OS cells (Bartkova et al., 2005). Strikingly, treatment of the HCT116-ER-E2F1 cells with OH-Tam results by itself in a rather weak increase of γH2A.X (Figure 4a) or phospho-ATM staining (Figure 4b), two components of DDR pathways. This indicates that oncogenic stress induced by nuclear E2F1 does not efficiently activate such pathways in these cells (see Supplementary data 5B and 5C for representative images). However, knockdown of p400 leads to a dramatic increase in γH2A.X or phospho-ATM-positive cells on OH-Tam treatment, indicating that knockdown of p400 restores the response to oncogenic stress. Moreover, this effect is completely lost in cells co-transfected by a Tip60 siRNA, stressing the importance of the p400/Tip60 ratio. Note that p400 knockdown is sufficient to induce a slight increase in γH2A.X staining in the absence of OH-Tam (Figure 4a), indicating that it favors a response to endogenous stresses in proliferative HCT116 cells, probably explaining its effects on proliferation and apoptosis induction described in the previous figures. Importantly, Annexin V analysis indicates that this DDR pathways activation by p400 knockdown is associated with increased apoptosis (Supplementary data 6). Taken together, these results indicate that p400 knockdown favors the Tip60-dependent DDR and the subsequent apoptosis induced by oncogenic stress, suggesting that the defect in the p400/Tip60 ratio observed in HCT116 cells is responsible for their lack of response.

Figure 4
figure4

p400 knockdown restores the response to oncogenic stress in HCT116 cells. (a) HCT116 cells stably transfected by the ER-E2F1 expression vector were treated or not with 300 nM hydroxytamoxifen (OH-Tam) for 48 h before to be transfected with the indicated siRNAs. After 96 h, cells were subjected to immunofluorescence staining using an anti-γH2A.X antibody. Shown is the quantification of a representative experiment, with cells separated in four groups according to the percent of the nucleus area that is stained (0–25, 25–50, 50–75 and 75–100%). (b) Same as in a using anti-phospho-ATM antibody and quantification of negative and positive cells.

Tip60 function is rate limiting for the response to 5-FU

Results from the above experiments underline the importance of the p400/Tip60 ratio in the response to stress. Because drugs used against cancer also induce stresses, we investigated whether the decrease in Tip60 function due to this defective p400/Tip60 ratio could be detrimental to the response of HCT116 cells to therapeutic drugs.

5-FU is an antimetabolite commonly used in first-line treatment of colon cancers. We found that Tip60 stable overexpression (using the cells described in Figure 2c) favors apoptosis induced by 5-FU, as indicated by FACS analysis (Figure 5a). These data indicate that Tip60 function is rate limiting for the response of HCT116 cells to 5-FU.

Figure 5
figure5

Manipulating the p400/Tip60 ratio affects HCT116 cells response to 5-fluorouracil (5-FU). (a) HCT116 cell population stably expressing Tip60 and a control cell population were treated 24 h with 5-FU and assessed for apoptosis induction by fluorescence-activated cell sorting (FACS) analysis of Annexin V/7-AAD staining. The mean and SDOM from three independent experiments are shown (relative to 1 in 5-FU-treated control cells). (b) HCT116 cells were transfected with the indicated siRNAs alone or in combination. After 48 h, cells were treated or not with 5-FU. After 24 h, apoptosis was assessed by Annexin V/7-AAD staining. The mean and standard deviation from three independent experiments are shown (relative to 1 in cells transfected with control siRNA and treated with 5-FU). (c) HCT116 cells were transfected with the indicated siRNAs alone or in combination. After 24 h, cells were counted and plated for clonogenic assay. At 24 h after seeding, 5-FU was added or not for 8 h. After 2 weeks, cells were stained with crystal violet and the number of clones was counted. The mean and standard deviation from three independent experiments are shown. A representative photography of cell dishes is also shown. Note that to calculate the mean and standard deviation from independent experiments, we standardized the number of clones relative to 1 for the number of clones in cells transfected by the control siRNA, both for untreated cells and for 5-FU-treated cells.

We then tested whether we can restore the response to 5-FU by manipulating the p400/Tip60 ratio. We found that transfection of p400 siRNA facilitates 5-FU-induced apoptosis (Figure 5b; see Supplementary data 7 for the absolute proportions of apoptotic cells). Moreover, co-transfecting a Tip60 siRNA abolishes the effects of p400 depletion. To investigate whether these changes in apoptosis induction can be translated into modifications in cell survival, we performed a clonogenic assay (Figure 5c). In untreated cells, we found that p400 inhibition decreases cell survival, in agreement with the fact that it induces apoptosis, whereas decreasing Tip60 levels has no effect. In cells treated with 5-FU, we found that the p400/Tip60 ratio is critical for cell survival, because decreasing Tip60 increases cell survival, whereas p400 siRNA transfection has an opposite effect. Moreover, the effects of p400 inhibition on cell survival are dependent on Tip60, as Tip60 siRNA co-transfection largely reverses the effects of p400. Taken together, these data underline the importance of the p400/Tip60 ratio for the response of HCT116 to 5-FU.

The importance of the p400/Tip60 ratio is not restricted to HCT116 cells

We next investigated whether the sensitivity to p400 inhibition is observed in other colon-cancer-derived cell lines than in HCT116 cells. We used the same siRNA-based strategy to alter the p400/Tip60 ratio. Importantly, we found that, in HT29 cells as in HCT116 cells, p400 and Tip60 do not regulate each other's expression (Figure 6a). TSA treatment of HT29 cells also induces apoptosis, which is significantly decreased by Tip60 siRNA (Figure 6b; P<0.029566). Note that Tip60 siRNA transfection inhibits apoptosis to a lesser extent in HT29 cells compared to HCT116 cells (Figure 3a), probably because of the relative weaker expression level of Tip60 in HT29 cells. Thus, in these cells also, the Tip60-dependent apoptotic pathway is functional. Modifying the p400/Tip60 ratio using siRNAs resulted in very similar effects than those observed for HCT116 cells: decreasing p400 levels led to increased apoptosis of HT29 cells both in the absence and in the presence of 5-FU (Figure 6c), which translated in decreased cell survival in a clonogenic assay (Figure 6d). Moreover, these effects are reversed by co-transfection of a Tip60 siRNA (Figures 6c and d). Thus the sensitivity to changes in the p400/Tip60 ratio is not restricted to HCT116 cells and is likely to be a feature of all—or a subset of—colon cancers. In particular, because HT29 cells are p53 deficient, contrary to HCT116 cells, this result indicates that p53 activity is not required for the decreased cell survival on p400 inhibition.

Figure 6
figure6

The p400/Tip60 ratio also controls the proliferation of HT29 cells. (a) HT29 cells were transfected with the indicated siRNAs and nuclear extracts were prepared 48 h later as described in Supplementary data 1; western blot analysis was then performed using antibodies directed against p400, Tip60 or HDAC1/2. The asterisk indicates a nonspecific band observed in the p400 western blot. (b) HT29 cells were transfected with the indicated siRNAs. After 36 h, trichostatin A (TSA, 400 ng/ml) was added or not. After 24 h, apoptosis was assessed by Annexin V/7-AAD staining. The mean and standard deviation from three independent experiments are shown (relative to 1 in TSA-treated cells transfected with control siRNA). (c) HT29 cells were transfected with the indicated siRNAs alone or in combination. After 48 h, cells were treated or not with 5-fluorouracil (5-FU). After 48 h, apoptosis was assessed by Annexin V/7-AAD staining. The mean and standard deviation from three independent experiments are shown (relative to 1 in cells transfected with control siRNA and treated with 5-FU). (d) HT29 cells were transfected with the indicated siRNAs alone or in combination. After 24 h, cells were counted and plated for clonogenic assay. At 24 h after seeding, 5-FU was added for 24 h. After 2 weeks, cells were stained with crystal violet and the number of clones was counted. The mean and standard deviation from three independent experiments are shown. Note that to calculate the mean and standard deviation from independent experiments, we standardized the number of clones relative to 1 for the number of clones in cells transfected by the control siRNA, both for untreated cells and for 5-FU-treated cells.

Discussion

Our results confirm our previous findings that p400 functions as an inhibitor of some Tip60-dependent pathways and that the effects of p400 knockdown are mediated through an increase in the activity of these Tip60-dependent pathways. Indeed, knockdown of p400 leads to same phenotypic consequences in colorectal-cancer-derived cell lines than overexpression of Tip60, that is, an increase in basal and 5-FU-induced apoptosis. Furthermore, all the effects of p400 knockdown are inhibited by the knockdown of Tip60. Interestingly, such a functional relationship between p400 and Tip60 could explain the link we observed between Tip60 and p400 mRNA expression changes in colon carcinoma: the decrease in p400 expression observed in some tumors would not be compatible with tumor proliferation (perhaps because of increased apoptosis due to activation of Tip60-dependent pathways), as suggested by our in vitro data. However, a concomitant decrease in Tip60 expression would inactivate these pathways and thus allow cells with decreased p400 expression to proliferate. By such a mechanism, cells with a decreased expression of both Tip60 and p400 would be selected. Note however that, although we show that Tip60 and p400 do not regulate each other expression, we cannot rule out the possibility that the correlation between Tip60 and p400 mRNA expression in colorectal cancers is due to co-regulation of their promoters.

The mechanism by which p400 knockdown can activate Tip60-dependent pathways is unclear for the moment. It has been proposed that p400 antagonizes Tip60 function in the p53 pathway by mediating, on p53-responsive promoters, the incorporation of H2A.Z, a histone H2A variant that antagonizes Tip60-activating abilities (Gevry et al., 2007). Clearly, our results indicate that the repressive effects of p400 on Tip60-dependent pathways are not restricted to p53-dependent promoters. Indeed, we found that p400 controls a DNA damage pathway upstream of p53 activation (actually upstream of ATM activation). Moreover, the Tip60-dependent apoptosis induced by p400 knockdown is also observed in p53-negative colon cancer cells (HT29 cells; Figure 6), indicating that p53 activity is not required for this effect. Importantly, we were able to rule out the possibility that p400 regulates Tip60 expression, at least in HCT116 and HT29 cells (see the western blots shown in Figures 3b and 6a). Because p400 physically interacts with Tip60 within the Tip60 complex, p400 could directly inhibit Tip60 histone acetyltransferase activity. In agreement with this hypothesis, we found that knocking down p400 increases global acetylation levels of histone H4, a known substrate of Tip60 (F Escaffit, L Mattera and D Trouche, unpublished results). However, because histone H4 acetylation is also induced by following induction of DNA damage (Murr et al., 2006), this finding could merely be an indirect consequence of the activation of DDR pathways by p400 knockdown. Strikingly, Tip60 directly acetylates the DNA damage signaling protein ATM (Sun et al., 2005), therefore allowing its activation. It is thus tempting to speculate that p400 directly or indirectly inhibits this acetylation event. Whatever the mechanism, our data indicate that, in colon cancer cells, the imbalance between p400 and Tip60 does not allow efficient activation of a DDR pathway on oncogenic stress.

Interestingly, a repressive role of p400 on DDR pathways could explain most, if not all, of the functions attributed to p400 so far. Indeed, knockdown of p400 function could lead to weak activation of DDR pathways in proliferating cells resulting in activation of the p53 pathway. In agreement with such a mechanism, we found that p400 knockdown was sufficient by itself to induce low levels of p53 phosphorylation in U2-OS cells (S Tyteca and D Trouche, unpublished results). Such an activation would result in senescence in untransformed cells (Chan et al., 2005) or cell-cycle arrest (Tyteca et al., 2006) or apoptosis (this study) in transformed cells. Also, by inducing DDR pathways, p400 knockdown could induce apoptosis in p53-negative cells through p53-independent pathways (see Figure 6).

Our results lead us to propose that p400 and Tip60 function, at least in part, as a couple with an effector (Tip60) and a regulator (p400). Strikingly, the ratio between these two proteins seems to be imbalanced in colon cancer, disfavoring some Tip60-dependent pathways. Moreover, Tip60 function is limiting for basal and 5-FU-induced apoptosis in colorectal-cancer-derived cell lines: overexpression of exogenous Tip60 or activation of endogenous Tip60 function favors apoptosis in a Tip60-dependent fashion. Taken together, these data suggest that a decrease of Tip60 function is likely to be causal for colon cancer progression by preventing apoptosis. In agreement with this possibility, a twofold decrease of Tip60, which is within the range of what we observe in colorectal cancer, favors tumorigenesis in a mouse model of lymphoma (Gorrini et al., 2007). This haploinsufficient tumor suppressor effect has been attributed to a function of Tip60 in the DDR pathway induced by oncogenic stress, which is a known anticancer pathway in mammals. In agreement with this possibility, we found that in colon-cancer-derived HCT116 cell line this pathway is defective but can be re-induced by inhibiting the expression of the Tip60 inhibitor p400. Thus, our data suggest that a decreased Tip60 function due to an imbalanced p400/Tip60 ratio favors progression toward colon cancer by preventing the activation of the oncogene-induced DDR pathways (see our model in Figure 7). More importantly, we also found that the decrease in Tip60 function is detrimental for the response of colon-cancer-derived cell lines to 5-FU, used as a first-line treatment against colon cancer. These results indicate that, if we are able to reestablish Tip60 activity in colorectal cancer, we might be able both to restore a certain level of apoptosis and to cooperate with conventional chemotherapeutic drugs.

Figure 7
figure7

Our working model of Tip60 function thresholds and cancer progression. In normal cells, Tip60 function allows cell proliferation and a normal response to stress (situation 1). On a slight reduction of Tip60 function (below threshold 1 but still above threshold 2) due to an imbalance in the p400/Tip60 ratio, Tip60 function is still sufficient to allow cell proliferation but cannot sustain a normal response to DNA damages: this situation favors cancer progression by inactivating anticancer barriers dependent on DNA damage response (DDR) pathways (situation 2). Importantly, Tip60 function cannot decrease below a second threshold (threshold 2), as this situation would not be compatible with cell proliferation (situation 3). Compounds that would restore the normal Tip60/p400 balance in cancer cells may allow Tip60 function to reach threshold 1, therefore activating DDR pathways. Such compounds may have therapeutic value.

Importantly, endogenous Tip60 is still functional in at least three established colorectal carcinoma cell lines (HCT116 (Figure 3a), HT29 (Figure 6b) and HCT8 (data not shown)), because activation of most acetylation-dependent pathways using the histone deacetylases inhibitor TSA leads to apoptosis in a Tip60-dependent way. These data are consistent with the idea that Tip60 activity should not decrease below a threshold level otherwise it would not be compatible with cell proliferation (most likely because Tip60 is important for the expression of genes essential for cell proliferation, such as E2F-dependent genes (Taubert et al., 2004)). Cancer cells would thus harbor a slightly decreased Tip60 function, high enough to be compatible with cell proliferation but not sufficient to promote stress-induced apoptosis (see the model in Figure 7). Our data obtained using p400 siRNA indicate that compounds that would interfere with the p400-mediated repression of Tip60-dependent pathways could be valuable drugs against colorectal carcinoma (in particular in combination with conventional chemotherapy), by restoring the correct balance between p400 and Tip60 and therefore reactivating stress–response pathways. Moreover, because Tip60 has been shown to be a haploinsufficient tumor suppressor in a wide variety of human cancers, it would be interesting to investigate the function of the p400/Tip60 ratio for cell proliferation and response to stress-inducing therapeutic drugs in other cancers.

Materials and methods

Antibodies, plasmids and siRNAs

The anti-Tip60 antibody was described previously (Legube et al., 2004). The anti-PARP antibody was purchased from Cell Signaling Technology (Danvers, MA, USA), anti-p400 from Abcam (Paris, France), anti-HDACs from Transduction Laboratories (Lexington, KY, USA), anti-HA from Covance (Madison, WI, USA), anti-γH2A.X from Cell Signaling Technology and anti-glyceraldehyde-3-phosphate dehydrogenase (GAPDH) antibody from Chemicon International Inc. (Temecula, CA, USA). All secondary antibodies were purchased from Amersham (Piscataway, NJ, USA).

pcDNA3-HA-Tip60 and retroviral pLPCX-Tip60 vectors were constructed by inserting the HA-tagged human Tip60 cDNA into pcDNA3 vector (Invitrogen SARL, Cergy Pontoise, France) or the wild-type human Tip60 cDNA into the retroviral pLPCX empty vector (Clontech Bio Europe, Saint-Germain-en-Laye, France). Details of constructions are available on request. siRNAs were described in Tyteca et al. (2006). Silencing efficiency was checked in each experiment by reverse transcription followed by QPCR as described (Tyteca et al., 2006).

Cell culture, drug treatment and siRNA transfections

Culture products were purchased from Invitrogen SARL. The GP293 retrovirus packaging cells were purchased from Clontech Bio Europe. HIEC and Caco-2 (clone 15) cells were kind gift from Dr JF Beaulieu (Université de Sherbrooke, Quebec, Canada). All cell lines and their derivatives were cultured in Dulbecco's modified Eagle's medium supplemented with antibiotics and 10% fetal calf serum (FCS), with the exception of HIECs (for which HEPES, 0.01 M, and epidermal growth factor, 1 ng/ml, were added) and Caco-2 cells (20% FCS). All drugs were purchased from Sigma-Aldrich (St Quentin, Fallavier, France). In experiments including 5-FU treatment, the drug was added in different amounts depending on cell line (for cytometry analysis: HCT116/100 μM, HT29/1 mM; for clonogenic assays: HCT116/1 mM, HT29/1 mM). For siRNA transfection, 5 × 106 cells were electroporated with 20 μl of siRNAs (100μM) using an electroporation device (Amaxa AG, Cologne, Germany), according to manufacturer's specifications.

Generation of stable cell lines

Colorectal cancer cell lines, control or overexpressing Tip60, were obtained using retroviral-mediated transfer as previously described (Escaffit et al., 2007) using pLPCX or pLPCX-Tip60 vectors. Puromycin (1 μg/ml; Invitrogen SARL) was added 48 h following infection and stable colorectal cancer RLPCXEmpty or RLPCXTip60 cell populations were obtained after 10 days of treatment.

A pCDNA-based expression vector encoding for HA-ER-E2F1 (kind gift from Professor K Helin) was transfected in HCT116 cells using Lipofectamine Plus reagent and puromycin was added for 10 days to obtain cells population stably expressing the fusion protein.

Western blot analysis

Nuclear extracts or total cell lysates were prepared as described in Supplementary data 1. The standard western blot procedure was used with LumiLight-plus reagent detection as previously described (Tyteca et al., 2006). Anti-HDACs or GAPDH antibodies were used as loading controls.

Growth suppression and clonogenic assays

For the growth suppression assay, cells were transfected using 10 μg of DNA by the calcium-phosphate method and were washed with Tris-buffered saline (50 mM Tris (pH 8.0), 150 mM NaCl) 16 h later. After 2 days, transfected cells were selected with G418 (800 μg/ml) during 3 weeks. Resistant clones were visualized with crystal violet coloration (Sigma-Aldrich).

Regarding the clonogenic assay, cells were electroporated with the indicated siRNAs. After 24 h, cells were counted and plated for clonogenic assay. At 24 h after seeding, 5-FU was added. After 2 weeks, cells were stained with crystal violet and the number of clones was counted.

Apoptosis assay using flow cytometry

Cells were harvested and treated using an Annexin V-fluorescein isothiocyanate/7-amino-actinomycin D (AAD) kit (Beckman Coulter, Marseille, France) according to the manufacturer's instructions. Cells were then analysed by flow cytometry.

DAPI staining, immunofluorescence and quantification

Immunofluorescence of cells on coverslips, observations and acquisition of native images were performed as previously described (Escaffit et al., 2007). Quantification of fluorescence levels was done using home-developed macros in ImageJ software (National Institutes of Health, Bethesda, MA, USA) to normalize background, thresholds and measures.

References

  1. Bartkova J, Horejsi Z, Koed K, Kramer A, Tort F, Zieger K et al. (2005). DNA damage response as a candidate anti-cancer barrier in early human tumorigenesis. Nature 434: 864–870.

  2. Berns K, Hijmans EM, Mullenders J, Brummelkamp TR, Velds A, Heimerikx M et al. (2004). A large-scale RNAi screen in human cells identifies new components of the p53 pathway. Nature 428: 431–437.

  3. Chan HM, Narita M, Lowe SW, Livingston DM . (2005). The p400 E1A-associated protein is a novel component of the p53 → p21 senescence pathway. Genes Dev 19: 196–201.

  4. Escaffit F, Vaute O, Chevillard-Briet M, Segui B, Takami Y, Nakayama T et al. (2007). Cleavage and cytoplasmic relocalization of histone deacetylase 3 are important for apoptosis progression. Mol Cell Biol 27: 554–567.

  5. Frank SR, Parisi T, Taubert S, Fernandez P, Fuchs M, Chan HM et al. (2003). MYC recruits the TIP60 histone acetyltransferase complex to chromatin. EMBO Rep 4: 575–580.

  6. Fuchs M, Gerber J, Drapkin R, Sif S, Ikura T, Ogryzko V et al. (2001). The p400 complex is an essential E1A transformation target. Cell 106: 297–307.

  7. Gaughan L, Brady ME, Cook S, Neal DE, Robson CN . (2001). Tip60 is a co-activator specific for class I nuclear hormone receptors. J Biol Chem 276: 46841–46848.

  8. Gevry N, Chan HM, Laflamme L, Livingston DM, Gaudreau L . (2007). p21 transcription is regulated by differential localization of histone H2A.Z. Genes Dev 21: 1869–1881.

  9. Gorrini C, Squatrito M, Luise C, Syed N, Perna D, Wark L et al. (2007). Tip60 is a haplo-insufficient tumour suppressor required for an oncogene-induced DNA damage response. Nature 448: 1063–1067.

  10. Ikura T, Ogryzko VV, Grigoriev M, Groisman R, Wang J, Horikoshi M et al. (2000). Involvement of the TIP60 histone acetylase complex in DNA repair and apoptosis. Cell 102: 463–473.

  11. Legube G, Linares LK, Tyteca S, Caron C, Scheffner M, Chevillard-Briet M et al. (2004). Role of the histone acetyl transferase Tip60 in the p53 pathway. J Biol Chem 279: 44825–44833.

  12. Lleonart M, Vidal F, Gallardo D, Diaz-Fuertes M, Rojo F, Cuatrecasas M et al. (2006). New p53 related genes in human tumors: significant downregulation in colon and lung carcinomas. Oncol Rep 16: 603–608.

  13. Miremadi A, Oestergaard MZ, Pharoah PD, Caldas C . (2007). Cancer genetics of epigenetic genes. Hum Mol Genet 16 (Spec No 1): R28–R49.

  14. Murr R, Loizou JI, Yang YG, Cuenin C, Li H, Wang ZQ et al. (2006). Histone acetylation by Trrap-Tip60 modulates loading of repair proteins and repair of DNA double-strand breaks. Nat Cell Biol 8: 91–99.

  15. Sun Y, Jiang X, Chen S, Fernandes N, Price BD . (2005). A role for the Tip60 histone acetyltransferase in the acetylation and activation of ATM. Proc Natl Acad Sci USA 102: 13182–13187.

  16. Sykes SM, Mellert HS, Holbert MA, Li K, Marmorstein R, Lane WS et al. (2006). Acetylation of the p53 DNA-binding domain regulates apoptosis induction. Mol Cell 24: 841–851.

  17. Tang Y, Luo J, Zhang W, Gu W . (2006). Tip60-dependent acetylation of p53 modulates the decision between cell-cycle arrest and apoptosis. Mol Cell 24: 827–839.

  18. Taubert S, Gorrini C, Frank SR, Parisi T, Fuchs M, Chan HM et al. (2004). E2F-dependent histone acetylation and recruitment of the Tip60 acetyltransferase complex to chromatin in late G1 . Mol Cell Biol 24: 4546–4556.

  19. Tyteca S, Vandromme M, Legube G, Chevillard-Briet M, Trouche D . (2006). Tip60 and p400 are both required for UV-induced apoptosis but play antagonistic roles in cell cycle progression. EMBO J 25: 1680–1689.

  20. Wilson AJ, Byun DS, Popova N, Murray LB, L'Italien K, Sowa Y et al. (2006). Histone deacetylase 3 (HDAC3) and other class I HDACs regulate colon cell maturation and p21 expression and are deregulated in human colon cancer. J Biol Chem 281: 13548–13558.

Download references

Acknowledgements

We thank Professor JF Beaulieu and K Helin for materials, Professor B Ducommun, Dr G Legube and Dr Y Canitrot for critical reading of the paper. We also acknowledge JJ Maoret, responsible person of the QPCR platform at IFR31 (INSERM), as well as S Mazeres (IPBS) and P Grosclaude (INSERM) for their help to format the data concerning tumor analysis. This work was supported by a grant to DT as a ‘subvention libre’ from the ARC and the Institut National du Cancer (INCa), Cancéropôle Grand Sud-Ouest (ACI 2004/2007, to CC). LM and ST were supported by studentships from the French Ministry of Science and the ARC, respectively, whereas FE was supported by a fellowship from the ‘Fondation de France’.

Author information

Correspondence to D Trouche.

Additional information

Supplementary Information accompanies the paper on the Oncogene website (http://www.nature.com/onc)

Supplementary information

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Mattera, L., Escaffit, F., Pillaire, M. et al. The p400/Tip60 ratio is critical for colorectal cancer cell proliferation through DNA damage response pathways. Oncogene 28, 1506–1517 (2009) doi:10.1038/onc.2008.499

Download citation

Keywords

  • colon cancer
  • apoptosis
  • chromatin
  • p400
  • Tip60

Further reading