Proapoptotic molecules directly targeting the BCL-2 family network are promising anticancer therapeutics, but an understanding of the cellular stress signals that render them effective is still elusive. We show here that the tumor suppressor p53, at least in part by transcription independent mechanisms, contributes to cell death induction and full activation of BAX by BH3 mimetic inhibitors of BCL-xL. In addition to mildly facilitating the ability of compounds to derepress BAX from BCL-xL, p53 also provides a death signal downstream of anti-apoptotic proteins inhibition. This death signal cooperates with BH3-induced activation of BAX and it is independent from PUMA, as enhanced p53 can substitute for PUMA to promote BAX activation in response to BH3 mimetics. The acute sensitivity of mitochondrial priming to p53 revealed here is likely to be critical for the clinical use of BH3 mimetics.
Major tumor suppressors pathways, such as these relying on pRB and/or protein 53 (p53), promote proapoptotic signals that ultimately converge on Mitochondrial Outer Membrane Permeabilization (MOMP).1 As a consequence, their loss in cancer cells results in failure to undergo MOMP in response to therapy, and approaches allowing to mitigate such defects are being actively investigated.
The BCL-2 (B-cell lymphoma/leukemia-2) family proteins are key regulators of MOMP and subsequent apoptosis.2, 3, 4 They are subdivided into three groups depending on their BCL-2 homology (BH) domain composition and their function: the multidomain anti-apoptotic proteins (BCL-2-like 1 (BCL-xl), BCL-2 and myeloid cell leukemia-1 (MCL-1), the multidomain proapoptotic proteins (BCL-2-associated X protein (BAX), BCL-2 antagonist/killer-1 (BAK)) and the BH3-only pro-apoptotic members (BCL-2-associated death promoter (BAD), Bcl-2-interacting mediator of cell death (BIM), BH3-interacting-domain death agonist (BID), NOXA and p53-upregulated modulator of apoptosis (PUMA)).5, 6, 7 Cell-fate decisions triggered by apoptotic stimuli are based on the relative amount of each BCL-2 protein as well as on the interplay between members of this family.5, 8, 9 One proximal step is the conversion of inert monomeric molecules of BAX/BAK into dimers that nucleate higher order oligomerization and lead to mitochondrial damage.10, 11, 12 This process of ‘activation’ can be induced by a subset of BH3-only proteins that directly interact with BAX/BAK (the so-called activators, BIM, BID and PUMA). Conversely, antiapoptotic proteins prevent this by interacting with BAX/BAK and/or activators.13, 14 This relies on the binding of the BH3 domain of the proapoptotic proteins to a hydrophobic cleft formed by the BH1-2 and -3 domains of BCL-2 homologs.15 This can now be pharmacologically modulated by ‘BH3-mimetics’ that target more or less selectively the BH3-binding pockets of BCL-2, BCL-xL or MCL-1.16
BH3 mimetics directly promote MOMP by releasing BH-3 activators and BAX/BAK from antiapoptotic proteins, hence their use may help restore apoptosis in cancer cells harboring defects in tumor suppressor pathways. However, tumor suppressors may provide additional cooperating signals that foster BH3 mimetic induced cell death, and whose absence may reciprocally limit BH3 mimetics efficiency. Consistent with the latter view, we recently showed that the pRB/E2F-1 pathway amplifies cell death induced by BCL-2/BCL-xL inhibition, by mediating caspase-dependent induction of the endogenous MCL-1 inhibitor NOXA.17 Likewise, p53, as a transcription factor, was shown to induce the expression of various apoptotic BCL-2 family genes18, 19 in addition to directly interacting with some BCL-2 family proteins.20, 21, 22, 23, 24, 25, 26, 27, 28 So far, no comprehensive study has investigated which, if any, of these effects may be critical to BH3-mimetic induction of cell death. We herein show that p53, even when expressed in viable, dividing cancer cells, promotes death signals that critically cooperate with BH3 mimetic treatment to trigger cell death.
Constitutive expression of p53 in HCT116 p21−/− cells contributes to induction of cell death by the BCL-2/BCL-xL inhibitor ABT-737
We have previously established that the colorectal cancer HCT116 p21−/− cell line is a model cell line that requires sustained inhibition of PUMA and BAX by BCL-xL to survive. This cell line is therefore a useful model to study the mechanisms leading to BAX-dependent cell death following BH3 mimetic inhibition of BCL-xL.13 Independently from p21 loss, the HCT116 p21−/− cells were shown to express constitutively high levels of p5329 (see also Figure 1a).
These cells are more sensitive than the parental (HCT116 wt) cell line to treatment with the BH3 mimetic inhibitor of BCL-2 and BCL-xL, ABT-737 (Figure 1b). Whereas down regulation of p21 by RNA interference in wild-type cells has no detectable impact on their resistance to ABT-737 (Figure 1c), downregulation of p53 in p21 null cells by a lentiviral-based shRNA significantly decreased their sensitivity (Figure 1d). Reciprocally, treatment of HCT116 wt cells with the Mouse double minute 2 homolog (Mdm2) inhibitor Nutlin-3a enhanced p53 expression and the response to ABT-737 (Figure 1e) whereas HCT116 p53−/− cells remained resistant to the combined treatment (data not shown). This effect was inhibited by down regulation of p53 (Figure 1f). Of note, p21 levels were induced by Nutlin-3a treatment in these assays, further underscoring the inconsequence of p21 expression on sensitivity to ABT-737 (Supplementary Figure S1A). Similar results were observed in the breast cancer Cal51 cell lines where ABT-737 and Nutlin-3a treatment synergized to induce p53-dependent cell death (Supplementary Figure S1B–C).
Altogether, these results show that p53, when expressed above a certain threshold, contributes to the apoptotic response induced by ABT-737.
p53 does not sensitize cells to ABT-737 through transcriptional induction of NOXA, PUMA or BAX
PUMA and BAX are major actors of the apoptotic response of HCT116 p21−/− cells to ABT-737, whereas BIM did not appear to contribute to these responses.13 Moreover, silencing of NOXA expression prevented cell death induced by ABT-737 (Figure 2a), indicating that this endogenous MCL-1 inhibitor has a role in cell death induction by BCL-2/BCL-xL inhibition. As NOXA, PUMA and BAX are transcriptional targets of activated p53,30 we investigated whether their constitutive induction by p53 prior treatment, and/or their induction by p53 during treatment, had a role.
NOXA expression was increased in HCT116 p21−/− upon ABT-737 treatment (Figure 2a), as reported in other cell lines.17, 31 Such induction was still detectable in p53-depleted cells (Figure 2b) and overexpression of NOXA in HCT116 p53-null cells neither induced cell death by itself nor sensitized them to ABT-737 treatment (Figure 2c). This argues that NOXA is an inducible actor of ABT-737-triggered apoptosis. However its induction does not absolutely require p53 and it is insufficient per se, in the absence of p53, to induce efficient cell death in combination with ABT-737.
The situation was different for PUMA and BAX. Indeed, investigation of kinetic changes in PUMA and BAX protein and mRNA expression levels during ABT-737 treatment brought no support to the hypothesis that their expression was induced at all (and a fortiori by p53) during treatment (Figure 3a). Moreover, PUMA and BAX were not detectably affected by silencing of p53 in HCT116 p21−/− cells, whether these were untreated or treated 24 h with ABT-737 (Figure 1d).
Indeed, ABT-737 had no effect on the expression of p53 itself in the p21−/− cells (Figure 3b) and the transcriptional activity of p53, as measured in a luciferase reporter assay, was not detectably modified by ABT-737 treatment in p21-null cells (Figure 3c). Importantly, neither Pifitrin-α (an inhibitor of p53-dependent transcriptional activation), nor α-Amanitin (an inhibitor of RNA polymerase-mediated transcription) inhibited cell death induced by ABT-737 in HCT116 p21−/− cells (Figure 3d) arguing that the transcriptional activity of p53 is dispensable for cell death induction by ABT-737 in these cells. Notably, we even noted a stimulatory effect of these compounds. This evokes a preceding publication who ascribed this enhancement to the fact that p53 transcriptional activity is not only dispensable for apoptotic responses, but also a restrain to the execution of p53 mitochondrial death program.32 Pretreatment with Pifitrin-α of the wild-type cells did not decrease cell death induced by the combined ABT-737 and Nutlin-3a treatment, indicating that the transcriptional activity of p53 is dispensable under these conditions (Supplementary Figure S1D).
Finally, it should be noted that the effects exerted by p53 are not functionally equivalent to those of the BH3-only proteins NOXA and/or PUMA. As shown below, p53 exerts apoptotic effects even in cells depleted of MCL-1 and thus they cannot solely rely on the induction of NOXA (which essentially functions as an inhibitor of MCL-1) or of NOXA-like molecules. Moreover, we found that in HCT116 p21−/− PUMA−/− cells, enhancement of p53 expression by Nutlin-3a treatment sensitized these otherwise resistant cells to ABT-737 treatment (Figure 4a and b). Cell death induced under these conditions was inhibited by silencing of p53 by RNA interference (Figure 4a) and by pretreatment with a broad-spectrum caspase inhibitor Q-VD-OPh (Figure 4b), whereas pretreatment with Pifitrin-α did not decrease the level of cell death (Supplementary Figure S1D).
Our data support the notion that p53 exerts an effect that adds to these of NOXA and PUMA to promote cell death in ABT-737 cells, and that this effect is in great part independent of its transcriptional activity.
p53 favors BAX activation in ABT-737-treated cells
As BAX has a major role in cell death induced by ABT-737 treatment or BCL-xL loss in HCT116 p21−/− cells,13 we investigated the influence of p53 on BAX ‘activation’ by ABT-737 treatment.
We first evaluated conformation changes in the amino-terminus domain of BAX by immunoprecipitation assays with the 6A7 antibody. 6A7 immunoreactive forms of BAX were detected in ABT-737-treated p21-null cells, but only weakly in wild-type cells and not in p53-null cells, indicating that the insensitivity of the wild-type and p53-null cells coincides with an absence of efficient activation of BAX (Figure 5a). The induction of the 6A7 positive form was severely impaired in p21-null cells depleted in p53 arguing that p53 favors conformation change amplitudes of BAX in response to ABT-737 (Figure 5a). Importantly, even though PUMA is a major driver of BAX ‘activation’ in ABT-737 treated HCT116 p21−/− cells,13 a significant conformation change in BAX was detected in HCT116 p21−/−PUMA−/− cells treated by the combination ABT-737/Nutlin3a (Figure 5b). This implies that enhanced p53 can substitute for PUMA to promote BAX activation and subsequent cell death (as shown in Figure 4).
Other features of BAX activation are mitochondrial translocation and oligomerization. To evaluate BAX mitochondrial translocation, percentages of cells with punctate BAX distribution were determined after ABT-737 treatment (Figure 5c and Supplementary Figure S2). BAX showed a punctate staining pattern that colocalized with a mitochondrial marker upon ABT-737 treatment in p21-null cells but not in cells infected with a p53 shRNA lentivector (Figure 5c). No translocation of BAX was observed in the wild-type cells and in p53-null cells, consistent with the absence of 6A7-positive BAX molecules in these cells (Figure 5c). In agreement with data above, oligomerization of BAX, assessed by crosslinking with disuccinimidyl suberate (DSS) was detected in ABT-737-treated p21-null cells, but only weakly in wild-type cells and not in p53-null cells. This oligomerization of BAX in ABT-737 treated HCT116 p21-null cells was abolished by silencing of p53 (Figure 5d).
p53 modulates both the inhibition of BCL-xL by ABT-737 and death signals downstream of BCL-xL
BAX activation is critically kept in check by the BH3-binding activity of BCL-xL in HCT116 p21−/− cells, as shown by the fact that inhibition of BCL-2/BCL-xL by ABT-737, and BCL-xL down regulation, both trigger BAX-dependent cell death.13 In further support to this, treatment of these cells with WEHI-539, a selective BCL-xL inhibitor, triggered p53-dependent cell death (Supplementary Figure S3). This indicates that p53 acts by favoring derepression of BCL-xL sensitive death signals by BH3 mimetics and/or by inducing death signals once BCL-xL is inhibited.
To investigate the former, we measured the spacial proximity between transiently transfected Luciferase-fused BAX and YFP-fused BCL-xL by bioluminescence resonance energy transfer (BRET) in live cells expressing p53 or not, treated or not with ABT-737. As previously reported,33 saturable BRET signals were observed between donor BAX and increasing levels of acceptor BCL-xL, indicative of a specific interaction between these proteins. Specificity was further confirmed by assessing that BRET signals were significanly less intense when a variant of YFP-BCL-xL carrying a single point-mutation in the BH3 binding site was used (Supplementary Figure S4). The saturation curves that were observed between BAX and wild-type BCL-xL in control HCT116 p21−/− and in p53-depleted cells were almost identical, suggesting that constitutive p53 does not impact on the propensity of BCL-xL to sequester BAX (Figure 6a). We then investigated the influence p53 might have on the sensitivity of these interactions to ABT-737. The effect of a range of ABT-737 concentrations on BAX/BCL-xL BRET signals in control and p53-deleted cells was evaluated. Silencing of p53 in p21-null cells consistently enhanced the resistance of BAX/BCL-xL interactions (evaluated by BRET) to ABT-737 (Figure 6b), indicating that constitutive p53 sensitizes BAX/BCL-xL interactions to ABT-737. Of note, p53 is not absolutely required for ABT-737 to inhibit BAX/BCL-xL interactions, as comparable inhibitions of BRET signals by concentrations above 1 μM of ABT-737 were measured in p53-proficient and p53-deficient cells. This may explain why we detected no difference in the effects of ABT-737 (used at 2 μM) on endogenous BAX/BCL-xL when we grossly evaluated them by co-immunoprecipitation assays in lysates from treated wild-type and p21−/− HCT116 cells (Supplementary Figure S5).
To evaluate whether p53 might exert an apoptotic effect beyond its modulation of BCL-xL sensitivity to ABT-737, we analyzed whether it would have a role even in cells deleted in BCL-xL. In agreement with previous results,13 silencing of BCL-xL in p21-null cells promoted a more efficient cell death in itself than ABT-737 treatment. This effect was not significantly enhanced by silencing of MCL-1 (Figure 7a), whereas silencing of MCL-1 strongly increased cell death of p21−/− cells treated with ABT-737 (Figure 7b). Silencing of the two antiapoptotic proteins lead to a weak level of cell death of HCT116 wt cells (Figure 7a) that was enhanced by Nutlin3a treatment (Figure 7d). Little cell death was measured in the p53 null cells in all these conditions (Figure 7a and c). Moreover, Pifithrin-α did not inhibit cell death in response to the silencing of BCL-xL and MCL-1 in p21 null cells (Figure 7e). Thus, p53 is required to promote cell death in cells in which the proapoptotic actors are no more sequestered by the anti-apoptotic counterparts, independently from its transcriptional activity.
As a whole, these data indicate that, in order to favor BAX activation and cell death in response to BH3 mimetics, p53 both facilitates derepression of BAX from BCL-xL by the BH3 mimetics and full-blown BAX activation once BCL-xL is inhibited (or it expression down regulated).
p53 heightens mitochondrial apoptotic priming
We inferred from the above results that, in addition to having a specific effect on the response of BAX/BCL-xL interactions to BH3 mimetics, constitutive p53 would exert a promiscuous effect that amplifies mitochondrial damage in response to most (if not all) perturbations of the BCL-2 network. To test this in an integrated manner, we performed BH3 profiling assays. In these assays, we evaluated, as a marker of mitochondrial integrity, the mitochondrial potential of permeabilized HCT116 p21−/− or p21−/− PUMA−/− cells (in which p53 was downregulated or not) incubated with a range of BH3 peptides and with ABT-737. As shown in Figure 8a, HCT116 p21−/− cells expressing PUMA and p53 (that is, infected with a control shRNA) showed dose-dependent response to BAX activating peptides (BIM-BH3 and BID-BH3) and the sensitizing peptides BAD-BH3 (which targets BCL-2/BCl-xL) and Harakiri (HRK)-BH3 (which only targets BCL-xL) but did not respond to the NOXA-BH3 peptide (which inhibits MCL-1). This is in agreement with a critical role for BCL-xL in maintaining the survival of these apoptosis competent cells. Downregulation of p53 by shRNA decreased the response to active peptides. Whereas p53-depleted cells had a diminished response even to the highest concentrations of BAD-BH3 and HRK-BH3, they only resisted to intermediate concentrations of BID and BIM-BH3 peptides (Figure 8a). This indicates that constitutive p53 may provide an active mitochondrial damaging signal that remains limiting when sensitizing peptides are used but that can be bypassed by high concentrations of activating peptides. Moreover, BH3 profiling experiments were consistent with the notion that this signal is independent from an effect of PUMA on mitochondrial priming (ref. 13 and manuscript in preparation). PUMA-deficient cells had, indeed, diminished responses to low doses of BIM-BH3 and to ABT-737 (compare Figure 8b with 8a), that were further decreased by p53 depletion (Figure 8b). Altogether, these data indicate that constitutive p53 heightens mitochondrial priming by an effect that is independent from, and adds to, that of PUMA.
ABT-737 and its orally available equivalent ABT-263 do not target MCL-1 but inhibit both BCL-2 and BCL-xL. As the latter is arguably the most potent anti-apoptotic protein of the family and a major contributor to drug resistance, such compounds are likely to have a role to play in a chemotherapeutic setting. The clinical use of BH3 mimetic inhibitors of BCL-xL is, however, limited by their secondary effects, in particular on platelets. Identifying parameters that determine the therapeutic window of these compounds is therefore important. It is thus of particular relevance to report that p53, even when expressed constitutively under conditions where it does not influence the expression of its proapoptotic transcription targets, enhances cell death induced by BCL-xL inhibition. On one hand, this implies that BH3 mimetics may not totally substitute for the lack of an active p53 tumor suppressor in cancer cells. On the other hand, it implies that healthy tissues may be more harmed than anticipated when BCL-xL inhibitors are combined with chemotherapeutic agents that even mildly affect p53.
BH3 mimetics, because they only inhibit selective subsets of antiapoptotic proteins, efficiently induce apoptosis when two types of signals are also present: signals that inhibit other anti-apoptotic proteins (MCL-1 in the case of ABT-737) and signals that trigger BAX activation. Our results are in line with preceding results and underscore that, to the very least in the case of ABT-737 treatment of the cells used here, NOXA contributes to the former signals, and constitutive PUMA to the latter. However, p53 does not promote ABT-737 induction of cell death solely via expression of either one of these two main actors. Indeed, p53 exerts a proapoptotic activity even in MCL-1-depleted cells. Thus, the apoptotic function of p53 extends beyond inhibition of MCL-1 by induction of NOXA. Moreover, p53 was dispensable for ABT-737 treatment to induce expression of NOXA, an effect we recently reported to rely on the pRB/E2F-1 pathway in p53 mutant cells.17 We also found no evidence of a role of p53 in the expression of PUMA in cells where it nevertheless contributes to cell death. The fact that enhancement of p53 expression sensitizes PUMA knock out cells to ABT-737 demonstrates that p53 exerts a BAX-dependent apoptotic effect that is independent from PUMA, even though it might cooperate with it to favor cell death.
That constitutive p53 heightens mitochondrial priming in cooperation with (and not via) BH3-only proteins is particularly well illustrated by BH3 profiling assays. These assays reveal that down regulation of p53 impacts on the mitochondrial response to numerous BH3 peptides, even in cells where this response is lowered by PUMA knock out (which impacts in itself on mitochondrial priming independently from p53). p53 depletion diminishes the response to low concentrations of activator BH3 peptides but not to higher concentrations. This is mostly consistent with the notion that constitutive p53 provides signals that cooperate with BH3-mediated activation of BAX, and that are dispensable for MOMP to occur when a sufficient amount of activator BH3 molecules are present. In agreement with this, we found that p53 was required for BAX translocation, change of conformation and oligomerization to be fully patent in cells treated with BH3 mimetics, even though all these events can be reproduced in the absence of p53 in cell-free assays.5, 13
Another important implication from the BH3 profiling assays is that the effect of p53 on MOMP lies, at least partly, downstream of anti-apoptotic proteins, as p53 depletion confers resistance to induction of MOMP by the highest concentrations of the sensitizer BAD and HRK-BH3 peptides we could test. In further support of this idea, we found that p53 still had an apoptotic role in cells that were depleted in BCL-xL and/or MCL-1. This implies that p53 would exert a promiscuous, positive, effect on BAX activity regardless of the identity of the antiapoptotic proteins that keep this activity in check. Our observations may thus extend to BCL-2 or MCL-1 inhibition. Arguably, the effect of p53 on the sensitivity of BCL-xL/BAX interactions to ABT-737 might appear as an additional effect, as it implies that p53 might have a specific influence on BCL-xL. We cannot formally rule out, however, that it ensues from a single effect on BAX that favors its displacement from BCL-xL in the same time as its activation.
Pretreatments with either Pifithrin-α (or α-Amanitin) did not inhibit cell death induced by ABT-737, or by the silencing of MCL-1 and BCL-xL, which is reminiscent of their lack of inhibitory effect on Nutlin-induced p53-dependent apoptosis in HCT116 cells.32, 34 This argues that p53 would exert at least some of its proapoptotic effects by a transcription-independent function,21, 23, 24, 26, 27, 35, 36, 37, 38 possibly from the cytoplasm where we found p53 to partly reside in the cells used here (data not shown). It is particularly relevant here to recall that the DNA binding domain of p53 can interact with BCL-xL. This interaction is favored by tetramerization of p53 and it influences the BH3 binding activity of BCL-xL even though it does not occur at the BH3 binding site stricto sensu.21, 23, 24 p53 might also directly activate BAX and or BAK by a similar, yet to be fully characterized, type of interaction.27, 28, 36, 37 A recombinant p53 constructs encompassing residues 79–39338 had no detectable effect on the priming of HCT116 p21−/− cells (whether they had been depleted in p53 by RNA interference or not prior the assays, data not shown). When incubated with mitochondria isolated from BAX/BAK null mouse embryonic fibroblasts, this recombinant p53 did not trigger cytochrome c release by itself, did not activate recombinant BAX, did not enhance the effects of tBID on BAX, and did not derepress the inhibitory effects of BCL-xL (Supplementary Figure S6). Since this construct lacks the N-terminal transcriptional activation region, this brings support to the notion that this latter portion, but not the DNA binding domain per se is involved in BAX activation.39, 40 Cis–trans interconversion of Pro47 in this domain by the prolyl isomerase Pin1 was shown recently to enhance p53 induced BAX activation.41 Of note, we could not find evidence of specific interactions between constitutive p53 and either BAX or BCL-xL in co-immunoprecipitations assays in the cells used in this study, and we failed to detect specific BRET signals between ectopic p53 and any of these BCL-2 proteins (data not shown). We thus can not formally rule out that p53, in addition to its direct effects on BAX, might exert other effects on mitochondrial apoptotic priming.
Taken together, our data unravel widespread effects of p53 on mitochondrial priming. Mitochondrial priming may thus acutely respond to the many stresses that stabilize and activate p53. As BH3 profiling assays are quite sensitive to detect such changes, that can occur without any overt activation of a p53 transcriptional program, this underscores their potential as biomarkers predicting cytotoxicity (and BCL-2 dependency) in a dynamic manner42 and the role of mitochondria as a «sounding board» of stress signals. From a clinical standpoint, the preferred sensitivity to BH3 mimetics of cells that express higher levels of p53 implies that BH3 mimetic treatment may skew clonal diversity and counter-select for cells with less (or less responsive) p53. Understanding the mechanistic basis for p53 induced priming, and finding approaches to bypass the consequences of p53 defects on this priming, are therefore required to avoid this bias.
Material and methods
The Cal51 cell line obtained from ATCC, was grown in RPMI medium. The HCT116 cell lines (parental, p21−/−, p53−/− and p21−/−PUMA−/−), grown in McCoy’s 5A, were kindly provided by Dr. B. Vogelstein (John Hopkins Kimmel Cancer Center, Baltimore, MD, USA). When specified, we used ABT-737 at 2 μM, WEHI-539 at 2 μM, Nutlin-3a at 10 μM (Sigma, St. Louis, MO, USA), Etoposide at 50 μM (Sigma), Pifithrin-α at 40 μM (Sigma), α-Amanitin at 10 μM (Sigma), QVD-OPh at 10 μM (R&D System, Minneapolis, MN, USA).
Plasmids, RNA interferences and recombinant proteins
Cells were transfected with pcDNA3 (as a control vector) or pcDNA3-NOXA vector using Lipofectamine2000 (Invitrogen, Carlsbad, CA, USA). Control siRNA or p21 siRNA (Santa-cruz Biotechnologies, Santa Cruz, CA, USA, respectively #44230 and #29427) were transfected using HiPerfect transfection reagent (Qiagen, Valencia, CA, USA). We used the following lentivirus: PFG12 and PLKO1; bacterial LacZ as control (5′-gtgaccagcgaatacctgt-3′), human BCL-xL (5′-AGGATACAGCTGGAGTCAG-3′), and human MCL-1 (5′-gaatgccagtgacctgtgt-3′); p53 and control were from Addgene, Cambridge, MA, USA (respectively #25636 and #8453). Lentivirus were engineered, produced and titrated as previously described.13 A multiplicity of infection of three was used, and further experiments were performed 2 days after infection. Cells infected by PLKO1 were selected by adding 1 μg/ml of puromycine to the medium, and cells were used during a maximum of 4 weeks.
The recombinant His-tagged full length BAX and caspase-8-cleaved BID (tBID) were prepared as previously described.43 The recombinant p53 was kindly provided by Professor Halazonetis and prepared as previously described.38
Bioluminescence Resonance Energy Transfert assays were performed as previously described.33 Briefly, the indicated cells were seeded on 12 plates and transfected with 50 ng/well of pRLuc-BAX vector (BRET donor, BAX fused to luciferase), and increasing amount (for saturation curves), or 200 ng/well (for dose-response curves), of peYFP-BCL-xL vector (BRET acceptor, BCL-xL fused to eYFP). Twenty-four hours later, cells were reseeded into white 96-well plate and treated as indicated. Light emissions at 485 and 530 nm were measured consecutively by using the Mithras fluorescence-luminescence detector LB-940 (Berthold, Thoiry, France) after addition of 5 μM of the luciferase substrate (coelenterazine H from Uptima, MontLucon, France). The BRET signal was determined by calculating the ratio of the light emitted by the acceptor protein eYFP-BCL-xL (530 nm) over the light emitted by the donor protein Rluc-BAX (475 nm), and corrected by subtracting the background signal detected with a non-tagged BCL-xL.
Mitochondrial profiling was performed as previously described.44 Briefly, 15 μl of BH3 peptide diluted in T-EB (300 mM Trehalose, 10 mM HEPES-KOH pH7.7, 80 mM KCl, 1 mM EGTA, 1 mM EDTA, 0.1% BSA, 5 mM succinate) were deposited per well in a black 384-well plate. Cells were washed in T-EB, resuspended and added to one volume of a dye solution containing 4 μM JC-1, 40 μg/ml oligomycin, 0.02% digitonin, 20 mM 2-mercaptoethanol in T-EB. This cell/dye solution stood 10 min at RT to allow permeabilization and dye equilibration. Fifteen microliters of the cell/dye mix was added to each treatment well of the plate (final concentration of 2 × 104 cells/well), shaken for 15 s inside the reader, and the fluorescence at 590 nm monitored every 5 min at 32 °C for 3 h. Percentage loss of Ψμ for the peptides is calculated by normalization to the solvent only control DMSO and the positive control FCCP. PUMA2A is an inert double alanine-substituted PUMA BH3 peptide serving as a negative control. Individual DBP analysis were performed using triplicates for DMSO, FCCP, and the different BH3 concentrations used, and the expressed values stand for the average of three different readings.
Luciferase assays were performed by cotransfecting cells with TA-luciferase control vector or p53-reporter luciferase vector (Affymetrix, Santa Clara, CA, USA, #LR0000 and #LR0057, respectively) and with pCMV-β-galactosidase vector. After treatment, cells were lysed under agitation in PLB (Promega, Madison, WI, USA) and luciferase activity was measured using Dual-Luciferase Reporter Assay System (Promega). β-galactosidase activity was measured using the specific assay reporter system (Promega). The luciferase activity was reported to the β-galactosidase activity, and is shown relative to the untreated condition.
Coimmunoprecipitation assays, using 1 μg of the indicated antibody and 500 μg of protein lysates in 1% CHAPS buffer with protease phosphatase inhibitors, were performed with protein A/G-agarose PureProteome system (Millipore) as previously described.13 CHAPS buffer has been previously reported not to activate BAX.45, 46 After western blotting, proteins were revealed using clean-blot IP reagent Pierce (Thermo Fisher Scientific, Rockford, IL, USA).
For immunocytochemical assays, cells were incubated 15 min with 150 nM de MitoTracker Red CMXRos (Invitrogen), washed with PBS and fixed with 1% paraformaldehyde in PBS for 30 min at RT. Cells were washed in PBS and permeabilized with 0.1% SDS in PBS for 10 min at RT. Cells were washed and, after a 30-min saturation with 5% BSA in PBS at 37 °C, were incubated with primary antibodies diluted in 1% BSA in PBS overnight at 4 °C (BAX A3533, Dako, Glostrup, Denmark). After three washes in PBS, incubation with secondary antibodies (anti-rabbit IgG coupled to Alexa Fluor 488) was performed in 1% BSA in PBS for 2 h at 37 °C. Cells were then washed and mounted in ProLong Gold antifade reagent (Invitrogen). Representative images were acquired on Zeiss Axiovert 200M.
After washing with PBS, cells for crosslinking experiments were lysed on ice using 1% CHAPS buffer with protease phosphatase inhibitors, and extract were cleared by centrifugation 15 min 13 000 g at 4 °C. Hundred micrograms of protein extract were incubated with 5 mM of DSS in Hepes 20 mM buffer. After 30 min agitation at room temperature, 50 mM of Tris-HCl pH7.5 was added to stop the reaction. After 15 min of incubation, crosslinked extracts were denaturated for electrophoresis and western-blotting.
Cell viability was determined by a trypan blue staining procedure.
Immunoblot and antibodies
After two washings with cold PBS, cells were lysed in ice-cold lysis buffer (SDS 1%; EDTA 10 mM; Tris-HCl pH8,1 50 mM; protease inhibitors, Na3VO4 1 mM, NaF 100 × ) and extracts were sonicated. Protein extracts were separated by SDS-PAGE, transferred onto a PVDF membrane (Millipore, Billerica, MA, USA) and revealed with a chemiluminescence kit (Millipore). Presented Western-Blot are representative of three independent experiments. Following antibodies were used: actin (MAB1501R, Millipore), β-tubulin (T0198, Sigma), BAX (A3533, Dako), BCL-xL (1018-1, Epitomics), cytochrome c (BD-Pharmingen San Diego, CA, USA), MCL-1 (sc-819, Santa Cruz Biotechnologies, Santa Cruz, CA, USA), NOXA (ALX-804-408, Enzo Life Science, New York, NY, USA), PUMA (12450, Cell Signalling), p21 (2947, Cell Signalling), p53 (#554294, BD-Pharmigen). The above antibodies against BAX, BCL-xL and p53 were also used for immunoprecipitations, as those against BAX 6A7 (ab5714, Abcam, Cambridge, UK) and FLAG-tag (F1804, Sigma).
Total RNA were extracted from cells using a kit Nucleospin RNA (Macherey-Nagel, Düren, Germany) following the manufacturer instructions. Retrotranscription was permormed from 500 ng RNA using Maxima First strand cDNA Synthesis kit for RT-q-PCR (Fisher Scientifique, Pittsburgh, PA, USA). Quantitative PCR was performed with EuroBioGreen qPCR Mix Lo-ROX (Eurobio, Courtaboeuf, France) from 4 ng of retrotranscripted RNA using a Stratagene Mx3005P (Agilent, Santa Clara, CA, USA). Data were analyzed using Pfaffl method with three references genes (ActB, RPLP0 and RPS18)47 and presented in mRNA levels relative to untreated cells.
Following primers were used:
Data were from three independent experiments. Unless otherwise stated, statistical analysis of data was performed using two-ways ANOVA with Bonferonni post-test on GraphPad Prism (La Jolla, CA, USA). Error bars represent S.E.Ms. The following symbols are used: *, ** and *** that correspond to a P-value inferior to 0.05, 0.01 and 0.001, respectively.
BCL-2-associated death promoter
BCL-2-associated X protein
B-cell CLL/lymphoma 2
BH3-interacting-domain death agonist
Bcl-2-interacting mediator of cell death
bioluminescence resonance energy transfer
myeloid cell leukemia-1
mitochondrial outer membrane permeabilization
Mouse double minute 2 homolog
p53-upregulated modulator of apoptosis
Lowe SW, Cepero E, Evan G . Intrinsic tumour suppression. Nature 2004; 432: 307–315.
Chipuk JE, Fisher JC, Dillon CP, Kriwacki RW, Kuwana T, Green DR . Mechanism of apoptosis induction by inhibition of the anti-apoptotic BCL-2 proteins. Proc Natl Acad Sci USA 2008; 105: 20327–20332.
Martinou J-C, Youle RJ . Mitochondria in apoptosis: Bcl-2 family members and mitochondrial dynamics. Dev Cell 2011; 21: 92–101.
Tait SWG, Green DR . Cell survival in tough times: The mitochondrial recovery plan. Cell Cycle Georget Tex 2010; 9: 4254–4255.
Chipuk JE, Moldoveanu T, Llambi F, Parsons MJ, Green DR . The BCL-2 family reunion. Mol Cell 2010; 37: 299–310.
Lalier L, Cartron PF, Juin P, Nedelkina S, Manon S, Bechinger B et al. Bax activation and mitochondrial insertion during apoptosis. Apoptosis Int J Program Cell Death 2007; 12: 887–896.
Shamas-Din A, Brahmbhatt H, Leber B, Andrews DW . BH3-only proteins: orchestrators of apoptosis. Biochim Biophys Acta 2011; 1813: 508–520.
Gautier F, Guillemin Y, Cartron PF, Gallenne T, Cauquil N, Le Diguarher T et al. Bax activation by engagement with, then release from, the BH3 binding site of Bcl-xL. Mol Cell Biol 2011; 31: 832–844.
Leber B, Lin J, Andrews DW . Still embedded together binding to membranes regulates Bcl-2 protein interactions. Oncogene 2010; 29: 5221–5230.
Czabotar PE, Westphal D, Dewson G, Ma S, Hockings C, Fairlie WD et al. Bax crystal structures reveal how BH3 domains activate Bax and nucleate its oligomerization to induce apoptosis. Cell 2013; 152: 519–531.
Moldoveanu T, Grace CR, Llambi F, Nourse A, Fitzgerald P, Gehring K et al. BID-induced structural changes in BAK promote apoptosis. Nat Struct Mol Biol 2013; 20: 589–597.
Peng R, Tong JS, Li H, Yue B, Zou F, Yu J et al. Targeting Bax interaction sites reveals that only homo-oligomerization sites are essential for its activation. Cell Death Differ 2013; 20: 744–754.
Gallenne T, Gautier F, Oliver L, Hervouet E, Noël B, Hickman JA et al. Bax activation by the BH3-only protein Puma promotes cell dependence on antiapoptotic Bcl-2 family members. J Cell Biol 2009; 185: 279–290.
Ren D, Tu HC, Kim H, Wang GX, Bean GR, Takeuchi O et al. BID, BIM, and PUMA are essential for activation of the BAX- and BAK-dependent cell death program. Science 2010; 330: 1390–1393.
Ding J, Mooers BH, Zhang Z, Kale J, Falcone D, McNichol J et al. After embedding in membranes antiapoptotic Bcl-XL protein binds both Bcl-2 homology region 3 and helix 1 of proapoptotic Bax protein to inhibit apoptotic mitochondrial permeabilization. J Biol Chem 2014; 289: 11873–11896.
Ni Chonghaile T, Letai A . Mimicking the BH3 domain to kill cancer cells. Oncogene 2008; 27 (Suppl 1): S149–S157.
Bertin-Ciftci J, Barré B, Le Pen J, Maillet L, Couriaud C, Juin P et al. pRb/E2F-1-mediated caspase-dependent induction of Noxa amplifies the apoptotic effects of the Bcl-2/Bcl-xL inhibitor ABT-737. Cell Death Differ 2013; 20: 755–764.
Levine AJ, Oren M . The first 30 years of p53: growing ever more complex. Nat Rev Cancer 2009; 9: 749–758.
Ryan KM, Phillips AC, Vousden KH . Regulation and function of the p53 tumor suppressor protein. Curr Opin Cell Biol 2001; 13: 332–337.
Comel A, Sorrentino G, Capaci V, Del Sal G . The cytoplasmic side of p53’s oncosuppressive activities. FEBS Lett 2014; 588: 2600–2609.
Deng X, Gao F, Flagg T, Anderson J, May WS . Bcl2’s flexible loop domain regulates p53 binding and survival. Mol Cell Biol 2006; 26: 4421–4434.
Green DR, Kroemer G . Cytoplasmic functions of the tumour suppressor p53. Nature 2009; 458: 1127–1130.
Hagn F, Klein C, Demmer O, Marchenko N, Vaseva A, Moll UM et al. BclxL changes conformation upon binding to wild-type but not mutant p53 DNA binding domain. J Biol Chem 2010; 285: 3439–3450.
Han J, Goldstein LA, Hou W, Gastman BR, Rabinowich H . Regulation of mitochondrial apoptotic events by p53-mediated disruption of complexes between antiapoptotic Bcl-2 members and Bim. J Biol Chem 2010; 285: 22473–22483.
Vaseva AV, Moll UM . The mitochondrial p53 pathway. Biochim Biophys Acta 2009; 1787: 414–420.
Chipuk JE, Kuwana T, Bouchier-Hayes L, Droin NM, Newmeyer DD, Schuler M et al. Direct activation of Bax by p53 mediates mitochondrial membrane permeabilization and apoptosis. Science 2004; 303: 1010–1014.
Leu J. I.-J., Dumont P, Hafey M, Murphy ME, George DL . Mitochondrial p53 activates Bak and causes disruption of a Bak-Mcl1 complex. Nat Cell Biol 2004; 6: 443–450.
Sot B, Freund SMV, Fersht AR . Comparative biophysical characterization of p53 with the pro-apoptotic BAK and the anti-apoptotic BCL-xL. J Biol Chem 2007; 282: 29193–29200.
Ferrandiz N, Martin-Perez J, Blanco R, Donertas D, Weber A, Eilers M et al. HCT116 cells deficient in p21Waf1 are hypersensitive to tyrosine kinase inhibitors and adriamycin through a mechanism unrelated to p21 and dependent on p53. DNA Repair 2009; 8: 390–399.
Oda E, Ohki R, Murasawa H, Nemoto J, Shibue T, Yamashita T et al. Noxa, a BH3-only member of the Bcl-2 family and candidate mediator of p53-induced apoptosis. Science 2000; 288: 1053–1058.
Song JH, Kandasamy K, Zemskova M, Lin Y-W, Kraft AS . The BH3 Mimetic ABT-737 Induces Cancer Cell Senescence. Cancer Res 2011; 71: 506–515.
Vaseva AV, Marchenko ND, Moll UM . The transcription-independent mitochondrial p53 program is a major contributor to nutlin-induced apoptosis in tumor cells. Cell Cycle Georget Tex 2009; 8: 1711–1719.
Bah N, Maillet L, Ryan J, Dubreil S, Gautier F, Letai A et al. Bcl-xL controls a switch between cell death modes during mitotic arrest. Cell Death Dis 2014; 5: e1291.
Arima Y, Nitta M, Kuninaka S, Zhang D, Fujiwara T, Taya Y et al. Transcriptional Blockade Induces p53-dependent Apoptosis Associated with Translocation of p53 to Mitochondria. J Biol Chem 2005; 280: 19166–19176.
Follis AV, Llambi F, Ou L, Baran K, Green DR, Kriwacki RW . The DNA-binding domain mediates both nuclear and cytosolic functions of p53. Nat Struct Mol Biol 2014; 21: 535–543.
Pietsch EC, Perchiniak E, Canutescu AA, Wang G, Dunbrack RL, Murphy ME . Oligomerization of BAK by p53 utilizes conserved residues of the p53 DNA binding domain. J Biol Chem 2008; 283: 21294–21304.
Nieminen AI, Eskelinen VM, Haikala HM, Tervonen TA, Yan Y, Partanen JI et al. Myc-induced AMPK-phospho p53 pathway activates Bak to sensitize mitochondrial apoptosis. Proc Natl Acad Sci USA 2013; 110: E1839–E1848.
Petty TJ, Emamzadah S, Costantino L, Petkova I, Stavridi ES, Saven JG et al. An induced fit mechanism regulates p53 DNA binding kinetics to confer sequence specificity. EMBO J 2011; 30: 2167–2176.
Haupt Y, Rowan S, Shaulian E, Kazaz A, Vousden K, Oren M . p53 mediated apoptosis in HeLa cells: transcription dependent and independent mechanisms. Leukemia 1997; 11 (Suppl 3): 337–339.
Chipuk JE, Maurer U, Green DR, Schuler M . Pharmacologic activation of p53 elicits Bax-dependent apoptosis in the absence of transcription. Cancer Cell 2003; 4: 371–381.
Follis AV, Llambi F, Merritt P, Chipuk JE, Green DR, Kriwacki RW . Pin1-induced proline isomerization in cytosolic p53 mediates BAX activation and apoptosis. Mol Cell 2015; 59: 677–684.
Montero J, Sarosiek KA, DeAngelo JD, Maertens O, Ryan J, Ercan D et al. Drug-induced death signaling strategy rapidly predicts cancer response to chemotherapy. Cell 2015; 160: 977–989.
Montessuit S, Somasekharan SP, Terrones O, Lucken-Ardjomande S, Herzig S, Schwarzenbacher R et al. Membrane remodeling induced by the dynamin-related protein Drp1 stimulates Bax oligomerization. Cell 2010; 142: 889–901.
Sarosiek KA, Chi X, Bachman JA, Sims JJ, Montero J, Patel L et al. BID preferentially activates BAK while BIM preferentially activates BAX, affecting chemotherapy response. Mol Cell 2013; 51: 751–765.
Hsu YT, Youle RJ . Nonionic detergents induce dimerization among members of the Bcl-2 family. J Biol Chem 1997; 272: 13829–13834.
Antonsson B, Montessuit S, Lauper S, Eskes R, Martinou JC . Bax oligomerization is required for channel-forming activity in liposomes and to trigger cytochrome c release from mitochondria. Biochem J 2000; 345 (Pt 2): 271–278.
Pfaffl MW . A new mathematical model for relative quantification in real-time RT-PCR. Nucleic Acids Res 2001; 29: e45.
We thank members of the ‘Cell Survival and Tumor Escape in Breast Cancer’ laboratory for their technical advice, fruitful comments and enthusiasm. We thank Thanos Halazonetis for his kind sharing of the p53 (79–393) construct. We thank Cécile Couriaud for her technical help in the preparation of lentivirus particles. We thank Philippe Hulin for his technical support at the Cellular and Tissular Imaging Core Facility of Nantes University (MicroPICell). JLP and CV are supported by grants from INSERM Region and the Ministère de la Recherche et de l’Enseignement Supérieur respectively. This work was supported by Région Pays de la Loire (CIMATH2), Ligue contre le Cancer (R13137) and ARC (R15083NN).
The authors declare no conflict of interest.
Edited by G Dewson
Supplementary Information accompanies this paper on Cell Death and Disease website
Rights and permissions
Cell Death and Disease is an open-access journal published by Nature Publishing Group. This work is licensed under a Creative Commons Attribution 4.0 International License. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in the credit line; if the material is not included under the Creative Commons license, users will need to obtain permission from the license holder to reproduce the material. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/
About this article
Cite this article
Le Pen, J., Laurent, M., Sarosiek, K. et al. Constitutive p53 heightens mitochondrial apoptotic priming and favors cell death induction by BH3 mimetic inhibitors of BCL-xL. Cell Death Dis 7, e2083 (2016). https://doi.org/10.1038/cddis.2015.400
This article is cited by
Partial p53 reactivation is sufficient to induce cancer regression
Journal of Experimental & Clinical Cancer Research (2022)
Bim, Puma and Noxa upregulation by Naftopidil sensitizes ovarian cancer to the BH3-mimetic ABT-737 and the MEK inhibitor Trametinib
Cell Death & Disease (2020)
p73 Regulates Primary Cortical Neuron Metabolism: a Global Metabolic Profile
Molecular Neurobiology (2018)
Curcumin suppresses gastric tumor cell growth via ROS-mediated DNA polymerase γ depletion disrupting cellular bioenergetics
Journal of Experimental & Clinical Cancer Research (2017)