Increased acetylation of lysine 317/320 of p53 caused by BCR-ABL protects from cytoplasmic translocation of p53 and mitochondria-dependent apoptosis in response to DNA damage
- 618 Downloads
Chronic myeloid leukemia (CML) is a disorder of hematopoietic stem cells caused by the expression of BCR-ABL. Loss of p53 has not been implicated as important for the development of CML. Mutations in p53 protein are infrequent, however they correlate with the disease progression. The absence of p53 mutations does not exclude the possibility that other dysfunctions play an important role in CML pathology. Acetylation represents a very potent posttranslational mechanism regulating p53 stability, transcriptional activity and localization. In this study we have investigated whether the expression of BCR-ABL could influence the acetylation of p53, specifically at lysine 317/320 (K317/K320), which has been shown to regulate nuclear export and transcription-independent apoptotic activity of p53. We found that BCR-ABL expression increases K317 acetylation of p53 and is able to prevent a drop in acetylation observed upon DNA damage, followed by translocation of p53 to the cytoplasm and by Bax activation. We have shown that this site plays a crucial role in the regulation of p53 localization and p53-dependent, Bax-mediated apoptosis. Our study presents a novel BCR-ABL-dependent mechanism protecting from DNA-damage-induced cell death. It can, in addition to already known mechanisms, explain the resistance to p53-dependent apoptosis observed in CML cells expressing wt p53. We propose that the acetyltransferases regulating the p53 acetylation could be an interesting and potent target for therapeutic intervention.
KeywordsBCR-ABL p53 Acetylation Apoptosis DNA damage PCAF
Activation of the DNA damage response is a crucial signalling pathway, which plays an important role not only in the chemotherapy-induced cell death but also protects from genomic instability which usually correlates with cancer development [1, 2, 3]. Cancer cells are often deficient in this signalling pathway due to mutations or deletion of p53, a critical regulator of the DNA damage response and DNA damage-induced apoptosis . In chronic myeloid leukemia (CML), a disorder of hematopoietic stem cells caused by the presence of BCR-ABL protein, mutations in p53 are associated with the disease progression and the blast crisis, however only in a minority of analysed cases [5, 6]. Despite infrequent mutations of p53 in CML, cells remain insensitive to apoptosis caused by DNA damage. Thus, it is possible, that other mechanisms inactivating the p53-dependent signalling are activated by BCR-ABL. The p53 protein is a crucial player in the regulation of cell cycle and cell death. It induces transcription of numerous cell cycle regulators and proapoptotic genes as well as represses the transcription of antiapoptotic proteins [7, 8, 9]. Alternatively, in a transcription-independent process, p53 promotes cell death by its translocation to the cytoplasm and mitochondria, which leads to the activation of proapoptotic Bax protein and permeabilization of the mitochondrial membrane [10, 11]. Posttranslational modifications, such as phosphorylation, acetylation, ubiquitination, methylation and sumoylation contribute to the activation of different p53-dependent responses [12, 13]. As presented recently by Yamaguchi et al. , p53 acetylation has been indicated as a crucial factor regulating the transcription-independent proapoptotic functions.
Unlike other lysines in the C-terminal p53 sequence, which are acetylated by the p300/CBP acetyltransferase, the Lysine 320 (K320) of human p53/317 (K317) of mouse protein seems to be the only one residue acetylated by PCAF, the p300/CBP-associated factor. However, it has been proposed that both, PCAF and p300 acetyltransferases physically interact and their relative contribution and specificity toward acetylation of substrates are still not clear . This lysine residue is located within a flexible linker domain near the oligomerization domain, which also contains a nuclear localization signal (NLS) and makes this residue crucial for the regulation of localization of the p53 protein [16, 17].
The influence of BCR-ABL oncoprotein on the acetylation of p53 has not been investigated intensively. As indicated by Lee et al.  cells expressing BCR-ABL show an aberrant protein acetylation status, however the exact specific sites were not studied. We used murine 32D cells stably expressing different levels of BCR-ABL to analyse whether BCR-ABL affects acetylation of lysine 317 of p53 and to study its role in the regulation of p53 localization and apoptosis upon DNA damage. We have found that BCR-ABL increases acetylation at the K317 residue of p53 in untreated cells as well as protects from deacetylation in response to DNA damage. Moreover, we have shown that the acetylation status of lysine 317 is crucial for regulation of the p53 nuclear-cytoplasmic shuttling and induction of the p53 and Bax-mediated cell death after DNA damage. These data reveal a novel BCR-ABL-mediated mechanism protecting from apoptosis. This mechanism, in addition to the other already known [19, 20], can promote the resistance to p53-dependent cell death activated by DNA damage occurring in CML cells expressing wt p53 protein. We propose that pathways participating in the acetylation of p53 at the K317/K320 residue in CML cells could be identified as prospective candidates for therapeutic targeting.
Materials and methods
Cell culture and treatment
32D mouse progenitor cells and C2 and C4 cell lines expressing low and high levels of BCR-ABL were kindly provided by Dr. S. McKenna (Cork Cancer Research Centre, Ireland) and maintained as described [21, 22]. Cells were treated with etoposide (Sigma-Aldrich) at a 7.5 μg/ml final concentration or etoposide together with 2 μM imatinib. C2 and C4 cells were incubated with etoposide and imatinib in medium containing IL-3. Imatinib was a generous gift from the Pharmaceutical Research Institute in Warsaw. PCAF small molecule inhibitor—anacardic acid derivative MG153 (6d)  was used at a 60 μM final concentration. Cells were pre-treated for 2 h with the inhibitor, followed by etoposide treatment. Pifithrin-μ (PFT-μ) (Sigma-Aldrich) was used at 1 μg/ml final concentration. Cells were pre-treated for 1 h, followed by etoposide treatment.
Cell lysis and immunoprecipitation
Cells were lysed with modified Ripa buffer [21, 22]. Equal amounts of the protein (750 μg) were taken for the immunoprecipitation. The supernatants were precleared by adding Protein A/G PLUS-Agarose Immunoprecypitation Reagent (Santa Cruz Biotechnology, Inc.) and incubated with “IP antibody–IP matrix complexes” overnight using p53 (1C12) mouse monoclonal antibody (Cell Signaling Technology) according to the protocol (ExactaCruz™ C, Santa Cruz Biotechnology, Inc.). Beads were washed with PBS with 5 μM trichostatin A and immune complexes were eluted with sodium dodecyl sulphate (SDS)-containing buffer and boiled.
Nuclear, cytosol and mitochondrial fractions
Nuclear, cytosol and mitochondrial fractions were prepared using kits: NE-PER® Nuclear and Cytoplasmic Extraction Reagents” (Pierce) and Mitochondria Isolation Kit for Cultured Cells” (Pierce).
Western blot analysis
For western blot analysis the following antibodies were used: p53 (1C12), phospho-p53 (Ser15), Hsp60 (D307), PCAF (all from Cell Signaling Technology), Bax (N-20) (Santa Cruz Biotechnology, Inc.), GAPDH (Abcam), PARP-1 (Alexis Biochemicals). To detect the level of K317 acetylation of p53, Anti-Mouse/Human p53 Acetylated Lys317 Polyclonal Antibody (PC-050, Trevigen, Inc.) was used. The ExactaCruz™ C-HRP antibody (ExactaCruz™ C Santa Cruz Biotechnology, Inc.) was used as a secondary antibody. Densitometric analysis of western blot bands for acetylated or total p53 and PCAF were done using the image station (GeneSnap from SynGene, Ingenious SynGene Bioimaging) and analysed using the density analysis software (GeneTools from SynGene). Densytometric analysis of acetylated p53 was normalized to total and the ratio Ac-p53 K317/total p53 was calculated.
Assessment of cell viability
Cell viability was assessed by the propidium iodide exclusion assay on FACS Calibur flow cytometer (Becton–Dickinson, Poland). Prior to the analysis cells were re-suspended in PBS with 50 μg/ml PI (Sigma).
Active Bax Form measurement
Cells were collected and fixed with 4 % paraformaldehyde. Staining was performed with the Anti-Bax Monoclonal Antibody 6A7 (BD Biosciences) that recognizes an epitope in the vicinity of the dimerization domain of Bax. Secondary antibody AlexaFluor 488 goat anti-mouse IgG (Molecular Probes, USA) was used at a concentration of 4 µg/ml. Samples were analysed on FACS Calibur flow cytometer.
Detection of γ-H2A.X
Staining was performed according to the procedure used for the measurement of the active form of Bax, using anti-phospho-Histone H2A.X (Ser139), FITC conjugate antibody (Millipore).
Mutagenesis and transfection
Plasmid containing full length cDNA of mouse transformation related protein 53 in pCMV-SPORT6 vector (Invitrogen), clone IRAVp968D0115D, was obtained from imaGenes GmbH. K317R substitution was introduced by amplification of full length plasmid with primers p53MmK317Rf (GAAAACCACTTGATGGAGAG) and p53MmK317Rr (TCTTTTGCGGGGGAG). PCR reaction was treated with DpnI to remove template DNA, 5′ ends were phosphorylated with T4 polynucleotide kinase and DNA was recircularized with T4 DNA ligase. Presence of substitution was confirmed by DNA sequencing. Cells were transfected using the Amaxa Nucleofector System, according to the protocol (Amaxa Nucleofector Technology); as a control, cells were transfected with the empty pcDNA 3.1 vector as well as pcDNA 3.1 vector containing wt p53. Cellular viability after nucleofection was routinely checked and the amount of dead cells was below 10 %. All treatments were done at 24 h after transfection.
PCAF siRNA (ON-TARGETplus smart pool, Mouse PCAF: ID-18519), p53 siRNA (ON-TARGETplus smart pool, Mouse TRP53: ID-22059) and negative siRNA (ON-TARGETplus Non-targeting siRNA #1) were obtained from Dharmacon. Transfection of siRNA (60 nM) was carried out using Amaxa Nucleofector System, according to the protocol (Amaxa Nucleofector Technology). Cells were treated with etoposide at 18 h after transfection.
Data are shown as mean ± SEM, of three independent experiments. The level of statistical significance was determined using the Student’s t test; *p < 0.05; **p < 0.005; ***p < 0.0005.
BCR-ABL expression prevents translocation of p53 from the nucleus to the cytoplasm and the mitochondria as well as Bax activation in response to DNA damage
First, to exclude the possibility that different sensitivity to apoptosis resulted from decreased level of DNA damage, we examined the presence of DNA breaks in etoposide-treated cells by detection of γ-H2AX histone using flow cytometry. We found the same level of γ-H2AX staining in all three cell lines (Suppl. Fig. S1a). Next, we assessed phosphorylation of p53 at the S15 and p53 stabilization upon DNA damage (Suppl. Fig. S1b and S1c) and did not find any differences between all cell lines, which indicated the activation of the DNA damage response. Increased expression of p21, a cell cycle inhibitor transcriptionally regulated in a p53-dependent manner, was detected in all of the cell lines (Suppl. Fig. S1c). Proapoptotic Bax protein was preferentially upregulated in 32D cells, as a result of inducing the p53-dependent apoptotic response (Suppl. Fig. S1c). Altogether, we showed that expression of BCR-ABL does not affect the DNA damage response and transcriptional activity of p53.
To directly study the localization of p53, we examined the level of p53 in the nuclear (lanes 1–4), cytosolic (lanes 5–8) and mitochondrial (lanes 9–12) fractions (Fig. 1c, upper panel). Contamination between fractions was excluded based on the analysis of specific markers (Fig. 1f). In all cells treated with etoposide, a strong induction of p53 in the nuclear fraction was detected, but only in 32D cells, the level of cytosolic p53 significantly increased, indicating translocation of p53 to the cytoplasm. Moreover, p53 was also found in the mitochondrial fraction of BCR-ABL-negative 32D cells. In C2 and C4 cells the increase of p53 in the mitochondrial and cytosolic fractions was nearly undetectable or not observed. This effect, visible only in C2 and C4 cells, was reversed by treatment with imatinib together with etoposide, indicating that BCR-ABL activity was responsible for altering the distribution of p53 (Fig. 1c, lower panel).
It is known that Bax protein must undergo conformational activation before it can be translocated to the mitochondria, which is mediated by the cytosolic pool of p53. It has been proposed that when p53 accumulates in the cytosol, it can function analogously to the BH3-only subset of proapoptotic Bcl-2 proteins to activate Bax and trigger apoptosis, however the direct mechanism of this activation is still not clear . Thus, to verify whether differences in the translocation of p53 will also correlate with different level of conformationally changed Bax, we performed flow cytometry analysis using an antibody recognizing only the active form of Bax (Fig. 1d). Cells expressing BCR-ABL did not show activated Bax upon etoposide treatment, in contrast to parental 32D cells, where a significant fraction of cells with conformationally changed Bax was detected. Treatment with etoposide together with imatinib reversed this protection and led to the activation of Bax protein also in C2 and C4 cells. Next, in order to check whether Bax translocated to the mitochondria, we studied the cellular fractions (Fig. 1e, upper panel). Bax translocation to the mitochondria in response to DNA damage was notable only in 32D cells. Retaining of Bax in the cytoplasm in C2 and C4 cells was a result of BCR-ABL activity, as after treatment with imatinib together with etoposide, the mitochondrial Bax was detected also in these cells (Fig. 1e, lower panel).
To prove that Bax expression in our model depended on p53 protein, 32D cells were transfected with siRNA against p53 (Suppl. Fig. S2). We found that silencing of p53 decreased Bax expression below the detection level. This data showed that in our system p53 protein was necessary for expression of Bax.
Based on the above results we postulated that inhibition of p53 translocation and p53-dependent apoptotic signalling might be the crucial mechanism contributing to resistance of CML cells to the DNA damage-induced apoptosis.
BCR-ABL expression leads to the increased acetylation of p53 at the K317 residue
We assumed that the altered acetylation of K317 residue might influence the subcellular localization of p53, and therefore we have analysed the levels of acetylated and total p53 upon etoposide treatment, followed by calculation of the ratio of Ac-p53 K317/total p53 (Fig. 3d). We found a decrease of acetylated p53 in etoposide-treated 32D cells (Fig. 3d, upper panel, lanes 1–3). This was not a result of the downregulation of the total p53 protein, as the level of p53 strongly increased upon etoposide treatment. Thus, the ratio of acetylated to total p53 drastically dropped in 32D cells and reached 0.2 of the control level (Fig. 3d, lower panel). Alternatively, in cells expressing BCR-ABL the amount of acetylated p53 remained unchanged, however a strong increase of total p53 was detected (Fig. 3d, lanes 4–9). Analysis of the ratio showed that in BCR-ABL-expressing cells the decrease of p53 acetylation upon DNA damage was transient and not as drastic as in 32D cells (Fig. 3d, lower panel). Following this transient decrease, which occurred after 3 h, the ratio parameter increased after 8 h and reached 0.6 and 0.8 of the control level in C2 and C4 cells, respectively. This indicated the existence of mechanisms, which protected p53 from deacetylation of K317 residue in cells expressing BCR-ABL.
Inhibition of BCR-ABL activity by imatinib in C2 and C4 cells treated with etoposide resulted in decreased acetylation of K317 residue of p53 (Fig. 3e, upper panel, lanes 4–9). Imatinib treatment did not affect the accumulation of the total p53 protein in all three cell lines. The ratio of acetylated to total p53 showed a strong drop not only in 32D cells, as observed before, but also in C2 and C4 cells, indicating that imatinib treatment diminished the mechanisms responsible for maintaining the acetylated status of p53 after DNA damage (Fig. 3e, lower panel).
Modulation of PCAF activity by small molecule inhibitor promotes translocation of the p53 protein to the cytoplasm and apoptosis after DNA damage
We found that similarly to C4 control cells, in cells treated with PCAF inhibitor as well as PCAF siRNA, p53 protein was upregulated in response to DNA damage, indicating that PCAF does not influence the accumulation of p53 (Fig. 4b).
Silencing of PCAF did not lead to translocation of p53 to the cytoplasm in response to etoposide (Fig. 4c). Also, no increase of apoptosis (Fig. 4d) as well as PARP cleavage (Fig. 4c) was observed. This data showed that either PCAF was not involved in the regulation of p53 translocation, which would be in contrast to current knowledge, or its function could be restored by other acetyltransferases, such as p300, which formed a complex with PCAF. It was also possible that the small amount of remaining PCAF protein efficiently protected from all visible effects.
In cells treated with MG153, which shows high specificity towards PCAF but also to some extent inhibits p300 acetyltransferase, a small fraction of cytosolic p53 was detected already in control cells, however a significant amount of p53 was translocated to the cytoplasm upon DNA damage (Fig. 4e). We also found that MG153 sensitized C4 cells to apoptosis induced by DNA damage and the level of cell death was significantly higher than detected in cells treated with the inhibitor alone (Fig. 4f). This result correlated with cleavage of PARP protein, which is a marker of apoptosis (Fig. 4e, middle panel). Altogether, these data suggest that p300/PCAF family of acetyltrasferases could play a role in the regulation of p53 translocation and sensitivity to DNA damage, however depletion of the PCAF protein alone did not cause any effect. As the main goal of this project was to precisely define the role of lysine 317 in the regulation of p53 localization, we performed more specific studies.
Acetylation of p53 at the K317 residue is a crucial factor regulating translocation of p53 to the cytoplasm and Bax-dependent apoptosis
To directly analyse the role of K317 acetylation, we used the acetylation defective p53 site mutant, where Lysine 317 was replaced by arginine (R), which similarly to lysine (K), is a basic and hydrophilic amino acid. This helped us avoid unspecific effects caused by the mutated form of p53. To compare, cells were also transfected with the vector carrying wt p53 as well as the control, insertless vector (pcDNA 3.1). C4 cells were transfected using Amaxa Nucleofector System with high efficiency (about 85 %). Nucleofection itself did not induce cell death and the viability after mock transfection was about 95 %.
As expected, we found that expression of the mutated p53 influenced neither the accumulation of p53 in response to etoposide (Fig. 5b) nor the DNA-damage response (Fig. 5c). Phosphorylation of p53 protein at S15 and the levels of target proteins transcriptionally regulated by p53-Bax and p21 were very similar in all three cell types—transfected with control vector (lanes 1–4), with the vector carrying the wt p53 (lanes 5–8) as well as the defective p53 (lanes 9–12). This proved that the K317R p53 mutant did not affect the DNA damage response, accumulation and transactivity of p53.
To check whether cytosolic translocation of p53 was followed by Bax activation, we investigated the presence of the conformationally activated form of Bax protein (Fig. 6b). We found an increased proportion of activated Bax in cells expressing mutated p53, in contrast to cells expressing wt p53. This was in line with our hypothesis that the cytosolic translocation of p53 was necessary for activation of Bax protein.
Next we studied whether unblocking of the p53-dependent events and Bax activation correlated also with mitochondrial localization of Bax (Fig. 6c). Bax protein was not found in the mitochondrial fraction of cells transfected with the empty vector or wt p53, however it was present in mitochondrial fraction of cells expressing K317R p53 mutant (Fig. 6c, lanes 7–9). Mitochondrial translocation of Bax is known to activate apoptotic response due to mitochondrial-dependent pathway. We found that expression of the mutated p53 significantly sensitized cells to etoposide (Fig. 6e), whereas in untreated cells apoptosis did not exceed 15 %. Our data clearly showed that the K317 residue of p53 played a significant role in the regulation of both, p53 localization and p53-dependent DNA damage-induced apoptosis. Moreover we showed that Bax-mediated apoptosis could be induced by cytosolic p53, however independently of mitochondrial p53.
Activation of the p53-dependent cellular response to genotoxic stress is frequently impaired in cancer cells. There is a growing body of evidence indicating that the posttranslational modifications of p53 play a crucial role in the regulation of p53-dependent pathways . It was recently shown that p53, in addition to its well known nuclear role as a transcription factor, is engaged in transcription-independent cytoplasmic and mitochondrial proapoptotic activities . Accumulation of p53 in the cytoplasm and mitochondria leading to apoptosis was shown in various experimental systems after exposure to DNA damaging drugs, irradiation, activated oncogenes and others [10, 31, 32]. As p53 is primarily a nuclear protein, the regulation of its nuclear export in response to apoptotic stimuli is one of the most important regulatory mechanisms. Additionally, according to current literature, sequestration of p53 in the cytoplasm is necessary for the activation of Bax protein, which also mediates apoptosis induction [24, 25, 33].
Our results presented in this paper have shown for the first time that expression of BCR-ABL oncoprotein increased acetylation of the K317 residue of p53. It is important that this increased acetylation protected from translocation of p53 to the cytoplasm upon DNA damage. We found that a block in the apoptotic signalling in response to DNA damage was caused by p53 export from the nucleus to the cytoplasm as well as by the activation and mitochondrial localization of Bax protein. Using PFT-μ we confirmed that selective inhibition of the transcription-independent branch of the p53 pathway is crucial to protect 32D cells from apoptosis. We also found that Bax translocation to the mitochondria required previous p53 activity.
This finding could be explained in the light of recent structural data on binding of PFT-μ to the p53–Bcl-XL complex resulting in inhibition of p53 translocation to the mitochondria . Authors provided the first evidence that PFT-μ binds not only to the p53-binding site, but also with the BH123 binding pocket, which forms the main interaction site and therefore regulates the MOMP and apoptosis. Thus, it binds the protein complex from both sites, protecting from p53 mitochondrial localization and inhibition of apoptotic signalling. It was proposed that when p53 accumulates in the cytosol, it can function analogously to the BH3-only subset of proapoptotic Bcl-2 proteins to activate Bax and trigger mitochondrial internalization followed by apoptosis . Thus, we suggest that PFT-μ, which has the ability to bind the p53-binding site and BH3 binding pocket, can protect not only from formation of the p53–Bcl-XL protein complex but probably also from formation of the p53-Bax transient complex . Altogether, this protects from mitochondrial sequestration of Bax and activation of the Bax-dependent death signal. Further, more direct, structural studies are necessary to fully explain this observation.
It was reported by other authors that Bax can induce apoptosis in the absence of p53 [35, 36, 37]. However, by blocking the ability of p53-dependent protein–protein interactions by PFT-μ, we excluded that Bax protein can be activated and mediate cell death in our system. Moreover, silencing of p53 decreased Bax protein below the detection level indicating that in our system expression of Bax depended fully on p53 transcriptional activity. Altogether we showed that p53 played an upstream regulatory role in Bax activation and induction of apoptosis.
Direct studies using acetylation defective K317R p53 mutant showed that the K317 residue of p53 played a significant role in the regulation of p53 nucleus-cytoplasm shuttling and transcription-independent apoptosis in leukemia cells expressing BCR-ABL. Mutated p53 was translocated to the cytoplasm but not the mitochondria; however it was still able to activate Bax in the cytoplasm. Activated Bax was conformatinally changed, which led to translocation to the mitochondrial membrane and apoptosis induction. This data clearly showed that Bax-mediated apoptosis could be induced by cytosolic p53 independently of mitochondrial p53.
We have also found that the mutation of K317 does not affect the stability of p53 as well as its transcriptional activity in response to DNA damage, indicating that lysine K317 is not required for these functions of p53. Our findings are supported by data presented by Chao et al.  clearly showing that abrogation of p53 acetylation at the K317 residue enhances the p53-mediated apoptosis after DNA damage, however without defects in p53 stabilization and activity. Similarly, mutation of this lysine to arginine disrupts the p53-mediated cell growth arrest and enhances the p53-directed apoptosis [16, 39].
It is important to emphasize that using the site specific p53 mutant we were able to separate two BCR-ABL-mediated mechanisms protecting from apoptosis caused by DNA damage: increased acetylation of p53, which influences the localization of p53 and Bax activation and translocation to the mitochondria, from mitochondrial sequestration of p53, which requires the ability to bind BCL-2 family proteins. In K317R mutants, p53 was not found in the mitochondrial fraction indicating that this event does not depend on acetylation of K317 residue of p53. Alternatively, both mechanisms were unblocked in imatinib and etoposide-treated cells, when BCR-ABL kinase activity was inhibited by imatinib. Imatinib treatment led to the translocation of both, Bax and p53 to the mitochondria in response to etoposide, confirming dependence on BCR-ABL activity. It was shown that Bax-independent mitochondrial apoptosis caused by p53 requires the interaction of p53 with the antiapoptotic proteins located at the outer mitochondrial membrane. Mihara et al.  first reported that the mitochondrial p53 preferentially engage Bcl-2 and Bcl-XL proteins, inhibiting their antiapoptotic function, thereby promoting mitochondrial membrane permeabilization. Alternatively, p53 can increase mitochondrial permeabilization by binding to the proapoptotic Bak protein. It was shown that BCR-ABL strongly increases the expression of Bcl-XL protein, which leads to resistance to DNA damaging agents. Imatinib downregulated its expression and sensitized cells to apoptosis . This data are consistent with our observations that apoptosis occurred when p53 as well as Bax were translocated to the mitochondria after imatinib and etoposide treatment.
Lysine 317/320, while acetylated, promotes nuclear localization, stabilization and an increased transcriptional activity of p53. As revealed by Bai and Zhu , upon acetylation, the DNA binding domain of p53 becomes active. It is already known that lysine 317/320 is subjected not only to acetylation but alternatively can also be competitively ubiquitynated and acetylation of C-terminal lysine residues of p53 inhibits their ubiquitination [43, 44]. Recently, mono-ubiquitynation of p53 has been proposed as a signal for nuclear export, leading to the accumulation in the cytoplasm and the mitochondria . We show that disruption of K317 acetylation of p53 leads to the cytoplasmic sequestration and activation of the transcription-independent apoptosis. It was also observed by others that forcing the cytoplasmic or mitochondrial localization of p53 can activate the p53-dependent apoptosis in vitro and in vivo [40, 46, 47].
There is a growing body of evidence concerning the role of PCAF and p300 acetyltransferases as well as their specificity in acetylation of different lysines of p53 protein . PCAF is proposed to specifically acetylate the K317 residue of p53 as we have already mentioned, however we did not detect any effect of PCAF silencing on the translocation of p53. Our data indicate that either PCAF is not involved in the regulation of p53 translocation in our system, which would be in contrast to current knowledge, or its function could be restored by other acetyltransferase—p300, which physically interacts and forms a complex with PCAF. It is also possible that the small amount of remaining PCAF protein efficiently protected from any visible effects. At this time we cannot explain this discrepancy. The system regulating acetylation of p53 is very complex and a separate study is necessary to distinguish the specific role of each acetyltransferase. However, we found that acetyltransferase inhibitor MG153 with high specificity to PCAF but probably able to inhibit p300 at least partially, overcomes resistance to DNA damage and induces p53 translocation and apoptosis in BCR-ABL-expressing cells, confirming the role of PCAF-related acetyltransferases in this processes.
In summary, we have provided evidence for a novel, BCR-ABL-mediated antiapoptotic mechanism, which results from the increased acetylation of p53 at the K317/K320 residue, thus preventing the nuclear export of p53 in response to DNA damage. Keeping in mind, that sequestrating of p53 to the cytoplasm and/or mitochondria has been proposed as an alternative therapeutic approach to induce death in cancer cells, our data clearly states that K317/K320 residue is an interesting site to be modulated in order to overcome resistance to apoptosis in CML cells. Although we are not convinced that PCAF alone is specifically responsible for regulation of p53 cytosolic translocation, we propose that targeting the mechanisms regulating acetylation of p53 using inhibitors, which usually have a broader spectrum of activities, should be considered as a prospective strategy to break resistance to DNA damage-induced apoptosis in CML cells.
This study was supported by a grant from Ministry of Science and Higher Education in Poland 2P04A 05729 (to K. P.). We thank the COST action Epigenetics: from bench to bedside. (TD0905) for financial support. We would like to thank Dr. E. Appella from NCI, NIH, Bethesda, for kindly providing a portion of Anti-Mouse/Human p53 Acetylated Lys317 Polyclonal Antibody (PC-050, Trevigen, Inc.).
Conflict of interest
The authors declare that they have no conflict of interest.
- 41.Horita M, Andreu EJ, Benito A, Arbona C, Sanz C, Benet I, Prosper F, Fernandez-Luna JL (2000) Blockade of the Bcr-Abl kinase activity induces apoptosis of chronic myelogenous leukemia cells by suppressing signal transducer and activator of transcription 5-dependent expression of Bcl-xL. J Exp Med 191(6):977–984PubMedCrossRefGoogle Scholar
- 42.Bai L, Zhu W-G (2006) p53: structure, function and therapeutic applications. J Cancer Mol 2(4):141–153Google Scholar