Tip60-mediated acetylation activates transcription independent apoptotic activity of Abl

Background The proto-oncogene, c-Abl encodes a ubiquitously expressed tyrosine kinase that critically governs the cell death response induced by genotoxic agents such as ionizing radiation and cisplatin. The catalytic function of Abl, which is essential for executing DNA damage response (DDR), is normally tightly regulated but upregulated several folds upon IR exposure due to ATM-mediated phosphorylation on S465. However, the mechanism/s leading to activation of Abl's apoptotic activity is currently unknown. Results We investigated the role of acetyl modification in regulating apoptotic activity of Abl and the results showed that DNA strand break-inducing agents, ionizing radiation and bleomycin induced Abl acetylation. Using mass spectrophotometry and site-specific acetyl antibody, we identified Abl K921, located in the DNA binding domain, and conforming to one of the lysine residue in the consensus acetylation motif (KXXK--X3-5--SGS) is acetylated following DNA damage. We further observed that the S465 phosphorylated Abl is acetyl modified during DNA damage. Signifying the modification, cells expressing the non acetylatable K921R mutant displayed attenuated apoptosis compared to wild-type in response to IR or bleomycin treatment. WT-Abl induced apoptosis irrespective of new protein synthesis. Furthermore, upon γ-irradiation K921R-Abl displayed reduced chromatin binding compared to wild type. Finally, loss of Abl K921 acetylation in Tip60-knocked down cells and co-precipitation of Abl with Tip60 in DNA damaged cells identified Tip60 as an Abl acetylase. Conclusion Collective data showed that DNA damage-induced K921 Abl acetylation, mediated by Tip60, stimulates transcriptional-independent apoptotic activity and chromatin-associative property thereby defining a new regulatory mechanism governing Abl's DDR function.

Consistent with their role in DDR, cells compromised for p300, Tip60 or hMOF function exhibit heightened sensitivity to DNA damaging agents due to defective cell-cycle checkpoint activation, DNA repair and or apoptosis [4,5]. Substrates of DNA damage-responsive acetyl-transferases include both histones and non-histone proteins such as p53, p73α, Ku70, E2-F1 and Sp1 [10][11][12][13][14][15]. In general, modification of histones by acetylation leads to unfolding of the nucleosomal fiber structure at DNA strand breaks perhaps facilitating access to DNA damage signaling and repair proteins [3,4]. On the other hand, acetylation of non-histone proteins is shown to stimulate their DNA-binding and transactivation function. In the case of p53, acetylation is shown to stimulate DNA-binding, stability and pro-apoptotic activity [10][11][12]. Similarly, acetylation of Ku70, E2F-1 and p73α selectively stimulates their apoptotic gene-specific transactivation function [13][14][15]. Interestingly, acetylation of Sp1 is shown to loosen bak promoter DNA binding [16]. Acetylation of ATM kinase, mediated by Tip60 complexed with MRE11-RAD50-NBS is shown to stimulate ATM's catalytic function during DNA damage [17]. Loss of acetylation due to depletion of hMOF by RNA interference prevents ATM-mediated phosphorylation of p53 and Chk2 [18]. Altogether, available evidences suggest that acetylation is an important modification that critically governs the activity of DDR proteins.
The ubiquitously expressed Src-related c-Abl tyrosine kinase is a well-documented regulator of cellular responses induced by oxidative stress, DNA damaging agents and growth factors [19][20][21][22]. Abl null mice display lymhopenia, reduced fertility, osteoporosis, and high neonatal mortality indicating its importance in normal growth and development [23]. Abl contains a large C-terminus whose deletion also caused neonatal lethality in mice signifying the biological role of this domain [24]. In this C-terminus, several discrete functional domains such as DNA binding domain (DBD), nuclear location and export sequences (NLS, NES) and RNA polymerase II-C-terminal binding domain (CTD) have been identified [25][26][27][28]. The presence of both NLS and NES is thought to allow nuclear-cytoplasmic shuttling of Abl and gain access to substrates in both compartments during signaling. The DBD of Abl is hyperphosphorylated by cdc2 during mitosis and this correlates with decreased chromatin association [26]. In vitro studies showed that Abl binds A/T-rich sequences through HMG-1 like boxes (HLBs) located within its DBD [27]. The RNAP II CTD is required for processive phosphorylation of Pol II largest subunit and for transcription enhancement function of Abl [28]. The catalytic function of Abl is normally tightly regulated through inter/intramolecular association [20][21][22] and phosphorylation [29] but stimulated upon exposure to IR or other genotoxic agents (e.g. cisplatin, methyl methane sulfonate, mitomycin C and hydrogen peroxide) [19,20]. Gamma-radiation induced Abl kinase activation is due to S465 phosphorylation mediated by ATM kinase [29]. Kinase-activated Abl, in turn, stimulates the transactivation property of p53 and its family members, p73α and p63α either through direct phosphorylation and/or association [30][31][32][33]. Corroborating well with attenuated induction or activation of Abl downstream targets, Abl -/mouse cells display apoptotic-resistant phenotype against a class of genotoxins [21,22]. Phosphorylation of a subset of DNA repair proteins, DNA-PK and RAD9 by Abl is shown to diminish their repair capability [34][35][36], although phosphorylation of RAD51 enhanced its association with RAD52 and consequently repair [37]. In sum, current data suggests that Abl modulates the cell death response induced by DNA damaging agents through multiple mechanisms involving activation of pro-apoptotic molecules and inactivation of DNA repair proteins.
We investigated whether acetyl modification is stimulatory for Abl's apoptotic function and observed that induction of DSB, either through IR exposure or treatment with belomycin, triggers c-Abl acetylation on K921. The requirement of S465 phosphorylation for K921 Abl acetylation further revealed that ATM-activated Abl is a substrate for DNA damage responsive acetyl-transferase/s. Attenuated apoptosis induced by DSB generating agents in cells expressing K921R mutant irrespective of new protein synthesis demonstrated the modification requirement in stimulating non-transcriptional apoptosis induced by Abl. Reduced chromatin binding displayed by K921R-Abl mutant compared to wild type Abl in IR exposed cells showed that acetylation positively modulates the DBD activity of Abl and may be closely linked to its nontranscriptional apoptotic activity. Finally, the ability of Tip60 to interact with Abl both in vitro and in vivo identified Tip60 as an upstream Abl modifier in the DNA damage signaling pathway.

IR exposure induces acetylation of c-Abl
To determine whether Abl is acetylated post DNA damage, extracts of HeLa cells exposed to either IR (5 Gy) or UV (20 J/m 2 ) were formed at 0, 2 and 4 h posttreatment and Abl was immunoprecipitated using anti-Abl (K12) antibody. Western blot analysis of the precipitants with pan-specific anti-acetyl lysine antibody revealed an immunoreactive band of~130 kDa corresponding to the size of Abl in IR exposed samples (Figure 1a; lanes 2 and 3, top panel). Neither untreated nor UV-treated extracts displayed this immune-reactive band despite comparable presence of Abl protein (bottom panel). A similar analysis using Abl -/-3T3 cells reconstituted with Flag-tagged Abl confirmed that the acetyl modification is induced specifically by IR but not UV (Figure 1b). We next radiolabeled Flag-Abl + 3T3 cells with [ 14 C] acetyl CoA and exposed to IR. Resolution of Abl immune-complex by SDS-PAGE followed by autoradiography showed a 14 C-labeled band of~130 kDa in IR exposed samples that coalesced with Abl band, revealed by immunoblotting (Figure 1c). Samples treated with UV did not display this radiolabeled band. To map the acetylation site, we transfected 293T cells with cDNA constructs of Abl containing truncations or internal deletions (Figure 1e), and then assessed for IRinduced acetyl modification. Wild type, ΔSca and ΔBgl-Abl showed acetylation whereas the DNA binding domain (DBD) deleted mutant constructs ΔXho and ΔXho-Sal did not display this modification (Figure 1d). The data showed that the DBD of Abl harbors the acetyl-modified residue/s.

K921 of Abl is acetylated in response to DNA damage
To identify the acetyl modified residues, 293 cells transiently expressing Flag-tagged Abl were exposed to IR and Abl was immunopurified, subjected to trypsin digestion, and analyzed by MS. Database searching using sequest identified mouse cAbl (P00520) with high confidence (Additional Files 1 &2). Specific MS/MS analysis of the peptide ion at m/z 477.8 corresponding to the (M+3H)3 + species showed a mass of 1431 Da, an increase of 56 Da compared to the calculated mass of peptide without modification. Acetylation of K921 and methylation of Cys899 (44Da + 14Da = 56Da) or carbamidomethylyation of Cys899 could account for the increased mass of the modified peptide; however, if Cys899 (57 Da mass Figure 1 IR exposure induces c-Abl acetylation. (a) HeLa cells were exposed to 5 Gy of IR or 20 J/m2 of UV and collected at 0, 2 and 4 h later. Anti-Abl immunoprecipitates were prepared using K-12 antibody and resolved on a 5-15% SDSPAGE. The proteins were transferred onto Immobilon-P and immune-analyzed with pan-specific anti-acetyl lysine antibody (top panel). The membrane was stripped as described in methods and reprobed with anti-Abl antibody (bottom panel). (b) Abl -/-3T3 cells reconstituted with Flag-tagged Abl were either mock, IR (5 Gy) or UV (20 J/m2) treated and collected 4 h later. Anti-Abl immunoprecipitates were prepared and immune-analyzed with acetyl lysine (top panel) or Flag antibody (bottom panel). (c) Abl -/-3T3 cells reconstituted with Flag-tagged Abl were labeled with 14C acetyl CoA for 2 h and then exposed to IR (5 Gy) or UV (20 J/m2) and collected 4 h later. Anti-Abl immunoprecipitates were prepared, resolved on 8% SDS-PAGE, and change) is carbamidomethylated, the corresponding Y 7 + , Y 8 + , and Y 9 + ions in the spectrum would be assigned to mass peaks that is approximately 1 Da higher in mass. MS/MS data, therefore, points to the possibility of acetylation on K921 with a better match of peaks between observed and calculated mass.
Located in the DBD (Figure 2a), K921 of Abl represents one of the two lysine residues within a consensus acetylation sequence (KAGK-PAQ-SPS) found in HAT substrates such as E2F-1, p53 and Ku70 [11][12][13][14]. As a logical extension, K921 and K924 of Abl were individually substituted with arginine (R) and tested for acetyl modification by IP followed by immunoblotting against anti-acetyl lysine antibody. K921R mutant was resistant to IR-induced acetyl modification demonstrated by lack of reactivity to acetyl-lysine antibody (Figure 2b). However, reactivity to K924-Abl remained comparable to WT. We next generated K921 site-specific acetyl-Abl antibody by immunizing rabbits with acetylated Abl peptide (described in Methods) and evaluated its reactivity to WT and K>R mutant of Abl by reprobing the membrane, shown in panel A. Result showed comparable reactivity to WT and K924R mutant of IR-treated cells but least to K921R mutant ( Figure 2c). We next immune-analyzed lysates formed from cells treated with DSB-inducing agent, bleomycin and observed acetylation of WT and K924R mutant but not K921R mutant at both 2 and 4 h post-treatment ( Figure 2d). Collective data established that DSB signals triggers K921 Abl acetylation.
S465 phospho-Abl is acetylated post DNA damage ATM-mediated phosphorylation on residue S465 stimulates Abl kinase activity upon IR exposure [29]. To study the interplay between kinase activation and acetylation, we tested the effect of Abl kinase inhibitor, imatinib (gleevec) on IR-induced K921 Abl acetylation. Immuneanalysis of the lysates with K921-Ac-Abl antibody revealed that the inhibitor blocked K921 Abl acetylation induced by IR (Figure 3a, compare lane 4 to 2). Reduced p-Tyr content on Abl in inhibitor-treated cells compared to control treatment, revealed by anti-p-Tyr (4G10) immunoblotting confirmed Abl kinase inhibition (bottom panel, compare lane 4 to 2). A similar result was obtained with the kinase-inactive Abl mutant (K290R) in that the mutant showed decreased K921 acetylation compared to WT after IR treatment ( Figure 3b). Examination of Abl acetylation in ATM-proficient (YZ-5) and deficient (EBS-7) fibroblasts showed that the modification induced by IR is dependent on ATM function ( Figure 3c, lane 2). To determine whether ATM phosphorylated Abl is subject to acetyl modification we next examined modification on the phosphorylation-resistant S465A and the constitutively active S465E mutant of Abl. Whereas WT and S465E-Abl mutant displayed K921 Abl acetylation, S465A mutant showed very little acetylation (Figure 3c, lanes 2 and 6). S465E mutant displayed acetylation in the absence of IR that was further enhanced after γ-irradiation by~5.53 fold (Figure 3d, lanes 5 and 6). Taken together, the results showed that ATM phosphorylated S465-Abl is acetyl modified on K921 post DNA damage.

Identification of Tip60 as a predominant Abl-specific acetylase
In an attempt to identify the acetylase responsible for modifying Abl, we initially examined the requirement of p300 HAT. However, comparable K921-Abl acetylation in HCT116/p300cells containing targeted deletion of exon 2 of EP300 gene [7] and in the parental HCT116 cell line demonstrated p300 dispensability (Additional File 3). We next tested the contribution of MYST-family acetyl-transferases, Tip60 and hMOF by transfecting HeLa cells with synthetic siRNA specific for either hMOF or Tip60. After confirming reduction of their expression by immunoblotting (Figure 4a), the transfectants were exposed to 5 Gy of IR and Abl acetylation was assessed. Result showed that the modification was partially inhibited in hMOFsuppressed cells but was completely abrogated in Tip60suppressed counter parts (Figure 4a, top panel, compare lane 4 and 6 to 2). To further evaluate the involvement of Tip60, we co-expressed Abl with either WT or HATdefective mutant of Tip60 (Q377E/G380E) in 293T cells and immune-analyzed the lysates for K921 Abl acetylation. WT Tip60 expressing population displayed modest acetylation that was greatly enhanced after γ-irradiation ( Similarly, anti-Tip60 immunoblotting confirmed the presence of equivalent levels of Tip60 in each of the HAT reactions (bottom panel). Together, the findings demonstrated that Tip60 acetylates Abl in response to DNA damage although a role for hMOF cannot be completely ruled out.

c-Abl co-precipitates with Tip60 acetylase following DSB induction
To examine whether c-Abl associates with Tip60 in vivo we performed co-immunoprecipitation assays using lysates prepared from Flag-Abl and HA-Tip60 transfected 293 cells. Result showed the presence of Abl in anti-Tip60 (HA) immunoprecipitates prepared from extracts of IR-but not mock-treated cells (Figure 5a). No association was observed in UV-treated cell lysates. When endogenous interactions between Tip60 and Abl was assessed in reconstituted Abl + 3T3 cells we observed that cells treated with bleomycin ( Figure 5b) showed enhanced association than observed with control treatment. The findings allow us to conclude that DNA damage enhances association between Abl and Tip60 consequently facilitating acetylation.

K921 acetylation is required for Transcriptionindependent apoptosis mediated by Abl
Given that acetylation activates the pro-apoptotic activity of E2-F1, p53 and p73α [12][13][14] we examined whether acetylation stimulates Abl's apoptotic function. For this purpose, cells reconstituted with WT or the non-acetylatable K921R-Abl mutant were exposed to 10 μΜ of belomycin and cell death was assessed at 48 h post-treatment. Annexin V/PI staining followed by flow cytometric analysis revealed~20.7% apoptosis (early and late) in cells expressing WT Abl (Figure 6a). Under similar conditions, % apoptosis in cells expressing K921R mutant was~9.6, a number that represents~2 fold reduction compared to WT. When the results of three independent experiments were combined, cells expressing the K921R mutant showed a statistically significant (*P < 0.01) reduced apoptosis compared to WT ( Figure 6b) consistent with acetylation as a required Figure 3 ATM phosphorylated Abl is targeted for acetyl-modification on K921. (a) HeLa cells were incubated with STI571 (10 μM; 1h) and then either mock treated or exposed to 5 Gy of IR. At 4 h post-treatment, lysates were prepared, adjusted for equal protein content and immune-analyzed with K921 Ac-Abl antibody (top panel) or K-12 Abl antibody (middle panel). Anti-Abl immunoprecipitates were also prepared and probed with anti-pTyr antibody (bottom panel). (b) IR-induced acetylation of WT and Kinase-defective Abl mutant. Abl-null 3T3s reconstituted with WT or kinase-defective Abl (K290R) were exposed to IR (5 Gy) and lysates were formed and subjected to immunoblot analysis with K921 Ac-Abl antibody. modification for activating Abl's cell death-inducing ability. To determine whether K921 acetylation promotes transcription-dependent or independent apoptosis, we included the de novo protein translation inhibitor, cyclohexamide in the assay. Although the inhibitor was slightly toxic demonstrated by high basal levels of cell death inWT and K921R mutant (~7.8% and 8.2%, respectively) it neither affected Abl expression nor K921 acetylation (Additional File 4). Nonetheless, when bleomycin-induced cell death was assessed in the presence of CHX, % cell death in WT-Abl cells was slightly reduced (~15.3%) compared to that observed in the absence (~20.7%). K921R mutant showed significantly reduced cell death both in the absence or absence of the inhibitor (~7.8% vs~9.6%). Attenuated apoptosis displayed by the K921R mutant irrespective of new protein synthesis demonstrated a requirement for Abl acetylation in triggering transcription-independent cell death. Correlating well these results, immunoblot analysis of the lysates prepared from these panel of cells with anti-PARP and anti-caspase-3 antibody (that recognizes cleaved fragment) showed comparable cleavage in bleomcyin-treated WT Abl cells in the presence or absence of CHX (Figure 6c, compare lane 2 and 4). Lysates of K921R mutant cells showed reduced but comparable cleavage irrespective of inhibitor treatment, consistent with attenuated cell death (compare lane 6 to 7).
To reinforce the results of the above analysis, HeLa cells stable depleted for Abl expression by shRNA (described in our previous publication Ref: 44) were transfected with WT or K921R mutant of Abl, along with transfection marker, GFP. At 36 h post-transfection, cells were exposed to 4.3 Gy of IR, a dose that triggers optimal cell death in these cell lines, and cell death was assessed by flow cytometric analysis of cells gated for GFP+ population (Figure 7a and 7b). Determination of % sub-G1 DNA containing population revealed reduced cell death in K>R mutant (~6.2%) compared to WT-Abl expressing cells (~14.6%). When cell death was assessed in the presence transcription inhibitor, α-amanitin, a RNA polymerase II poison, we observed slightly reduced cell death in WT-Abl cells (~11.6%) compared to control treatment (~14.6%). However, K921R mutant showed similar levels of cell death both in the presence and absence of the inhibitor (~6.2% vs~5.3%). Consistent with these results, comparably diminished PARP and caspase-3 cleavage was observed in lysates formed from K921R expressing cells with or without α-amanitin (Figure 7c). Taken together, the findings reinforced the involvement of Abl, and more importantly, K921 Abl acetylation requirement in promoting transcription-independent apoptosis triggered by DSB-inducing agents.

K921 acetyl-Abl is enriched in chromatin fraction of IRtreated cells
Abl association with chromatin is negatively regulated during mitosis by cdc2-mediated phosphorylation on multiple ser/thr residues, located in the DBD of Abl [26]. To determine whether Abl acetylation on K921 affects its chromatin-associative property, we compared the presence of acetyl-Abl in chromatin fractions of mock and IR exposed cells. Extracts formed from mockand IR-treated Abl + 3T3 cells were separated into nuclear and cytoplasmic fractions as described in Methods and their integrity was confirmed by immunoblotting with chromatin marker, Orc2 and cytoplasmic marker, tubulin (Figure 8a). Nuclear fractions were then divided into equal aliquots and immunoprecipitated with either K921 Ac-Abl or Abl antibody. Anti-Abl immunoblotting showed increased presence of both acetyl-Abl and total Abl in chromatin fractions of IRtreated extracts compared to mock-treated samples (top panel, compare lane 2 to 1). Cytoplasmic fractions showed no alterations in Abl content between mockand IR-treated cells (bottom panels, lanes 5 and 6). modification positively modulates Abl's chromatin-binding activity and may be closely linked to its non-transcriptional apoptotic activity.

Discussion
Induction of apoptosis is a central function of c-Abl in the DNA damage signaling pathway; however, the mechanism/s leading to activation of its apoptotic activity is not known. While previous studies established that the kinase function of Abl is essential for its cell death-inducing function [20][21][22] and is activated by phosphorylation, the present study demonstrates that acetylation activates Abl's apoptotic activity. Specifically, our study shows that K921 of Abl, located within the DNA binding domain, and confining to one of the lysine in the consensus acetylation motif identified in pro-apoptotic molecules such p53, E2F-1 and p73α [12][13][14][15] is modified by acetylation during DNA damage. Acetylation selectively activates the nontranscriptional cell death induced by Abl as the non-acetylatable K921R mutant displayed attenuated apoptosis compared to WT in the absence of new protein synthesis. Decreased chromatin binding displayed by the acetyl-resistant K>R mutant compared to wild-type supports the argument that acetyl modification positively modulates chromatin-binding property of Abl and may be closely linked to its apoptotic activity. Finally, our observation that Tip60 associates with and acetylates Abl both in vitro and in vivo identified Tip60 as a predominant DNA damage responsive acetylase that modifies Abl on K921.
Results of the present study raise an interesting question about how DNA damage induces K921 acetylation and activates its apoptotic activity. Acetylation of pro-apoptotic molecules such as p53 and p73α activates selective transactivation function towards apoptotic genes, bax and puma through enhanced promoter DNA binding [11,15]. On the other hand, acetylation of Sp1 is shown to loosen binding to bak promoter consequently allowing access of Sp3 to the promoter region to drive bak transcription [16]. However, unlike p53, p73α and Sp1, c-Abl is not known to bind specific DNA sequence. Instead, in vitro DNA binding studies showed that Abl binds A/T rich DNA sequences through its HMG-1 like boxes (HLBs), located within its DBD [27]. David and coworkers [38] further reported that Abl binds DNA sequences containing AAC motifs and to distorted DNA structures. HMG proteins such as HMG1A and HMGI/Y are shown to bind A/Trich stretches of chromatin through AT-hooks and contribute to chromatin remodeling [39][40][41]. Moreover, HMG proteins have been reported to displace DNA repair molecules from chromatin allowing the access of DNA caspases and nucleases [40][41][42]. Phosphorylation of a subset of DNA repair proteins, Rad9 and DNAPK by Abl is documented to suppress their repair capability [34,36]. When taken together, the findings lead us to speculate that acetylated Abl induces transcription-independent apoptosis through binding to AT/AAC rich sequences in chromatin facilitating remodeling and or inhibiting DNA repair. Another intriguing possibility is that Abl may promote apoptosis through increased Bad/Bax protein accumulation. In the cytoplasm, pro-apoptotic proteins such as Bax and Bad are kept inactive through interactions with 14-3-3 proteins. Yoshida et al [43] showed that JNK-mediated phosphorylation of 14-3-3 dissociates 14-3-3/c-Abl complex consequently leading to increased nuclear accumulation of c-Abl. Conceivably, nuclear retention of acetyl modified Abl through chromatin association may lead to reduced availability of Abl in the cytoplasm consequently allowing accumulation of Bad and Bax in the cytoplasm and enhancing cell death. Additional studies, however, are required to fully uncover the precise mechanism by which Abl promotes transcriptional-independent apoptosis. The mechanism by which DNA damage induces Abl acetylation by Tip60, at present, remains speculative. The lack of acetyl modification on DNA binding domain deleted mutants of Abl (ΔXho-Sal and ΔXho) indicates a requirement and an involvement of this domain for the modification. In addition, Abl requires ATM phosphorylation and residue (S465) for the modification since the phosphorylation resistant mutant, S465A-Abl failed to undergo acetylation upon exposure to IR and bleomycin. Moreover, the kinase-defective K290R-Abl was resistant to acetylation triggered by these DSB-inducing agents. We speculate that Abl either binds DNA lesion directly through its DBD or recruited to the lesion as part of ATM complex for interaction and acetylation by Tip60.Acetylation does not influence Abl kinase activity since activation was observed within 1 h post IR and acetylation was observed 2 h later. Moreover, mutants of Abl that cannot be activated (S465A-Abl and K290R-Abl) are resistant to acetylation induced by IR. The constitutively active S465E-Abl mutant, on the other hand, displayed K921 acetylation that was further enhanced after irradiation. These observations suggest that acetylation is a secondary modification that occurs on kinase-activated Abl.
Although Tip60 is essential for K921 Abl acetylation, the current study does not preclude hMOF acetylase in mediating the modification since hMOF-suppressed cells showed partially abrogated acetylation. Cross-talk between Tip60 and human MOF has been reported [9]. Moreover, hMOF has been reported to influence ATM activation in response to DNA damage as depletion of hMOF prevents ATM-mediated phosphorylation of p53 and Chk2 [18]. Such findings impart a role of hMOF in Figure 8 Enrichment of K921 Acetylated Abl in chromatin fractions of g-irradiated cells. (a) Reconstituted Abl + 3T3 cells were exposed to IR (5 Gy) and at 18 h post-treatment, cells were harvested, lysed in hypotonic buffer and fractionated into nuclear and cytoplasmic fraction as described by Méndez et al [46]. Nuclear fraction was again separated into chromatin bound and unbound and subjected to immunoprecipitation with anti-Abl (K-12) antibody. The immune-complex was analyzed by subjecting to western blotting with K921 Ac-Abl or Abl antibody (top panels). Input lysates were also blotted with tubulin (cytoplasmic marker) or ORC2 (chromatin marker) antibody (bottom panel). (b) Abl -/-3T3 cells reconstituted with K921R mutant of Abl were exposed to 5 Gy of IR and nuclear and cytoplasmic fractions were formed and immune-analyzed for K921 Ac-Abl, Abl, tubulin or ORC2 proteins as described in panel a.
acetylating Abl in response to DNA damage, in addition to Tip60. It is interesting to note that pro-apoptotic proteins such as p53 and E2F-1 are initially modified by phosphorylation, and at latter time-point, by acetylation. Whereas phosphorylation of p53 leads to stabilization, enhancement of promoter DNA binding and activation of its cell-cycle arrest function, acetylation is shown to positively modulate its apoptotic activity. Clearly, a parallel can be drawn between Abl and these DDR proteins in terms of regulation of their activity by coupled process of phosphorylation/acetylation. Specifically, when taken with the previous findings that observation that Abl contributes to growth arrest response [45], and is activated by ATM-mediated phosphorylation upon irradiation [29] lead us to speculate that combined modification by phosphorylation and acetylation modulates the cell growth and apoptotic activity of Abl following DNA damage.
Significance of the current study can be realized from the critical role that Abl plays in mediating the apoptotic response to chemotherapeutic agents that inflict genome damage [20][21][22]. A recent study showed that activated endogenous Abl kinase dramatically promotes breast cancer cell proliferation and anchorage-independent growth in serum, as well as survival following nutrient deprivation [46]. Additionally, as a co-activator of Androgen receptor (AR), Tip60 contributes to the progression of chemotherapeutic-resistant prostate cancer cell proliferation [47]. Given these observations, the identification of Abl as a Tip60 substrate may be a useful therapeutic tool for developing novel therapeutics in the treatment of cancer.

Conclusion
In summary, our results reveal a new regulatory mechanism controlling the apoptotic activity of Abl kinase during DNA damage. Specifically, our data shows that K921 Abl acetylation, mediated by Tip60, is a relevant posttranslational modification that stimulates the transcription-independent apoptosis induced by DSB. The positive influence exerted by acetylation on chromatin binding function of Abl suggests a link to promoting its apoptotic activity independent of transcription.

Construction of c-Abl mutants and Site-directed mutagenesis
Construction of murine c-Abl (type IV) internal and truncated mutants was described in a previous study [28]. Site-specific Abl mutations (K921R, K924R) were created using PCR-based Quickchange kit from Stratagene. The oligonucleotides and its complementary reverse oligo used as primers for K921R is 5'g cctcctcctgc ctgcacagggcgcgcaggcaagcccgca cag agc3'/5' gctctgtgcgggc ttgcctgcgcg ccc tgtgcaggcaggaggagg c 3'. For K924R mutagenesis: 5' cctgcctgcacaggaaaagcagggcgccccgca cag agc ccc agccaa g 3'/5' c ttggctggggctctgtgcggggcg ccc tgcttttcctgtgca ggcagg3'. PCR was carried out using primers (0.1 μmol/L) of primers, 2.5 units Taq DNA polymerase, and 3 to 15 ng plasmid DNA. The 1.9-kb BssH1-Kas1 fragments containing the desired mutations were subcloned into pCDNA 3.1 minus vector containing full-length WT Abl cDNA. All constructs were verified by direct DNA sequencing.

Mass spectrometry
293 cells transfected with Flag-tagged Abl (mouse Type1V) cells were exposed to 5 Gy of IR and harvested 4 h later. Cells were lysed in 1× lysis buffer (described above) and incubated with the anti-Flag monoclonal antibody-conjugated M2 agarose beads (Sigma). The bound material was resolved on a 5-15% SDS-PAGE and the Abl band corresponding to~130 kDa was excised, incubated with trypsin, and the products were run through a Thermo Finnigan LCQ DECA XP Plus ion trap mass spectrometer interfaced with a Shimadzu Binary HPLC to carry out liquid chromatography-tandem mass spectrometry (LC-MS/MS) analysis.

HAT Assays
Anti-Tip60 immune-complexes were obtained by incubating lysates formed from mock and irradiated HeLa cells with 5 ul of the antibody as described above. The immunecomplex was washed twice in HAT assay buffer (50 μΜ Tris, pH 8.0/10% glycerol/0.1 μΜ EDTA/1 μΜ DTT) and incubated with GST-Abl fusion proteins (0.5 μg) in the presence of 0.5 μCi [acetyl-1-14C] (370 kBq) and cold Acetyl CoA (10 μM). The reaction was carried out in 60 μl of HAT buffer for 30 min at 30°C and terminated using 3× SDS sample buffer. Reaction mixture was resolved on a 5-15% SDS-PAGE, transferred onto Immobilon-P and exposed to autoradiography after flurography.

Chromatin Fractionation
Chromatin fractionation was performed as described by Mendez and Stillman [48] with minor modifications. Approximately 1 × 106 cells were transfected with either Flag-tagged WT or K>R mutant of Abl using 3T3 Transfection reagent from Mirus, WI. At 36 h post-IR, cells were exposed to 5 Gy of IR and collected 4 h post-treatment. Cells were washed with PBS and resuspended in 200 μL of Buffer A containing 10 μΜ Hepes, 10 μΜ KCl, 1.5 μΜ MgCl2, 0.34 M sucrose, 10% glycerol, 1 μΜ NaF, 1 μΜ Na2VO3, 0.1%Triton X-100, and protease inhibitors. Cells were incubated on ice for 5 minutes, and cytoplasmic proteins were separated from nuclei by low-speed centrifugation at 1,200 ×g for 5 minutes. Nuclei were washed with Buffer A and lysed in 200 μL of Buffer B containing 3 μΜ EDTA, 0.2 μΜ EGTA, 1 μΜ DTT, and protease inhibitors followed by incubation on ice for 10 minutes. The lysate was centrifuged at 1,700 × g for 4 min. to separate soluble nuclear proteins from chromatin. The isolated chromatin was washed with Buffer B and pelleted by centrifugation at 10,000 ×g for 1 minute. Chromatin was resuspended in SDS-sample buffer and sheared by sonication and resolved on 5-15% SDS-PAGE. The proteins were transferred onto Immobilon-P and western blotted against the indicated antibody.