Histone Deacetylase Inhibitor LAQ824 Both Lowers Expression and Promotes Proteasomal Degradation of Bcr-Abl and Induces Apoptosis of Imatinib Mesylate-sensitive or -refractory Chronic Myelogenous Leukemia-Blast Crisis Cells

INTRODUCTION

The reciprocal chromosomal translocation t (9;22)(q34;q11) creates the Philadelphia chromosome and the bcr-abl fusion gene, which is the molecular, hallmark of 95% of CML 2 (1 , 2) . The dysregulated activity of the TK encoded by the bcr-abl importantly contributes toward the malignant phenotype of Bcr-Abl-expressing leukemia cells in the chronic and advanced blastic phases of CML (1 , 2) . For example, Bcr-Abl TK activates the PI3K/AKT kinase pathway resulting in the increased expression of cyclin D2 (3 , 4) , as well as the phosphorylation, proteasomal degradation, and decreased levels of the cyclin-dependent kinase 2 inhibitor p27, thus promoting cell cycle progression and cell proliferation (5 , 6) . Also, Bcr-Abl-expressing leukemia blasts of CML-BC display differentiation arrest and resistance to apoptosis, even after exposure to high doses of antileukemia drugs (7 , 8) . Intracellularly, Bcr-Abl activates several diverse molecular mechanisms known to inhibit apoptosis, which contribute toward the drug resistance of the leukemia cells (1 , 2 , 9) . These include increased phosphorylation and transactivation by signal transducers and activators of transcription 5, which leads to increased expression of the antiapoptotic Bcl-x_L protein (10, 11, 12) . Bcr-Abl TK also activates PI3K/Akt kinase and nuclear factor-κB signaling, which activate several antiapoptotic mechanisms (13, 14, 15, 16) .

Because of the proliferation and survival advantage conferred by Bcr-Abl, depleting its levels or activity has emerged as a rational strategy to treat the chronic and advanced phases of CML (9 , 17 , 18) . STI-571, now known as Gleevec (imatinib mesylate), is a phenylaminopyrimidine derivative and a relatively specific ATP binding site antagonist of Bcr-Abl TK (19) . At relatively low and clinically achievable concentrations (∼0.25 μm), imatinib selectively causes in vitro and in vivo growth arrest and apoptosis of Bcr-Abl-positive leukemia but not of normal hematopoietic progenitor cells (20, 21, 22, 23) . Imatinib has demonstrated an impressive clinical activity in the chronic phase of CML, with ∼95% of patients achieving durable complete clinical remissions (24) . However, ∼50% of patients with the chronic phase fail to achieve a cytogenetic complete response after imatinib therapy (24 , 25) . In the accelerated phase of CML and CML-BC after treatment with imatinib, the complete clinical and cytogenetic remission rates are even lower, with most patients suffering relapse and succumbing to imatinib-refractory CML-BC (24 , 26 , 27) . The mechanisms of resistance to imatinib identified in CML-BC cells include mutations in the kinase domain of bcr-Abl and amplification of the bcr-abl gene, resulting in the overexpression of Bcr-Abl (28 , 29) . Importantly, ectopic expression of the mutant Bcr-Abl was shown to transform murine hematopoietic cells and confer resistance against imatinib (28 , 30) . Recently, in 29 of 32 patients, whose disease relapsed after an initial response to imatinib, 15 different amino acid substitutions affecting 13 residues were identified as mutations in the Bcr-Abl kinase domain (30) . Two groups of mutations were noted. These included those that altered amino acids that directly contact imatinib (e.g., T315I) or those that prevented Bcr-Abl from achieving the inactive conformational state required for binding to imatinib (e.g., E255K). These mutations conferred varying degrees of resistance to imatinib in the leukemia cells. These observations emphasize the need to develop and test novel anti-Bcr-Abl agents, which have mechanisms of antileukemia activity distinct from those described for imatinib. Especially attractive would be those agents that not only enhance the antileukemia effects of imatinib but also exert cytotoxic effects against imatinib-refractory CML-BC (9) .

Histone acetyltransferases and HDACs are able to regulate gene expression patterns through recruitment to target genes in association with specific transcription factors (31) . HDACs catalyze deacetylation of lysine residues in the NH₂-terminal tails of the core nucleosomal histones, resulting in chromatin condensation and transcriptional regulation of cell cycle- and differentiation-regulatory genes (32) . Consequently, treatment with HDIs cause hyperacetylation of the NH₂-terminal lysine residues of the nucleosomal histones, which transcriptionally up-regulates p21^WAF1 expression and induces cell cycle arrest and apoptosis of cancer and leukemia cells (33 , 34) . Recently, a natural product HDI, depsipeptide (FK228), was also shown to cause acetylation of the hsp90. This was shown to destabilize the stable chaperone complex of hsp90 with its client proteins, thereby promoting their degradation by the proteasome. These client proteins include ErbB2, Raf-1, and mutant p53 (35) . A different class of HDIs, the hydroxamic acid analogues, e.g., TSA and SAHA, have also been demonstrated to exert potent in vitro and in vivo antileukemia effects (36, 37, 38) . Recently, a cinnamic acid hydroxamate, LAQ824, has been shown to act as a potent HDI at submicromolar levels (39) . In the present studies, we determined the effects of LAQ824 treatment on the mRNA and protein expression of bcr-abl. We also determined the effects of LAQ824 on the cell cycle status and apoptosis of the cultured and primary CML-BC cells. Whether treatment with LAQ824 would deplete the mutant Bcr-Abl and induce apoptosis of imatinib-refractory human CML-BC cells was also probed in the present studies. Finally, because it had been previously reported that Bcr-Abl is a client protein for hsp90 (40) , we determined whether LAQ824 treatment would induce acetylation of hsp90 and whether this would disrupt the chaperone association of Bcr-Abl to hsp90, leading to the proteasomal degradation and depletion of Bcr-Abl.

MATERIALS AND METHODS

Reagents.

LAQ824 and imatinib were kindly provided by Novartis Pharmaceuticals, Inc. (East Hanover, NJ). The pyrido[2,3-d] pyrimidine derivative PD180970 was synthesized by Parke-Davis Pharmaceuticals and was kindly provided by Dr. Alan Kraker (41) . PD180970 was diluted in DMSO. Anti-Bcl-x_L was purchased from PharMingen, Inc. (San Diego, CA), anti-Abl antibody from Santa Cruz Biotechnology (Santa Cruz Biotechnology, CA) Anti-Hsp70 and anti-Hsp90 antibodies were purchased from StressGen Biotechnologies Corp. (Victoria, British Columbia, Canada). Polyclonal anti-PARP antibody was purchased from Cell Signaling (Beverly, MA). The antibodies for the immunoblot analyses to detect the levels of proteins, including Bcl-xL, phospho-AKT, AKT, p21, and p27 were obtained, as described previously (38 , 40 , 42) .

Primary and Cultured CML-BC Cells.

CML-BC K562 and LAMA-84 cells were cultured as described previously (42) . The primary CML-BC cells were procured from the peripheral blood and/or bone marrow of 5 patients who had CML-BC. Of these, 3 patients showed clear evidence of disease progression in the bone marrow and/or peripheral blood while receiving imatinib at a dose of 600 mg/day. Two additional samples were purified from the peripheral blood and/or bone marrow obtained from imatinib-naïve patients. All samples were obtained with informed consent as part of a clinical protocol approved by the Institutional Review Board. Bone marrow aspirate or the peripheral blood sample was obtained in heparinized tubes, and mononuclear cells (BMMCs) were isolated by centrifugation through Ficoll-Hypaque (Amersham Pharmacia Biotech) and washed with RPMI 1640. To isolate CD34+ leukemic blasts from BMMCs, the cells were coated with anti-CD34-PE conjugated antibody (PharMingen, Inc.) and anti-PE microbeads (Miltenyi Biotech). The CD34+ cells were then isolated using magnetic cell sorting with LS+ columns and Variomax magnets (Miltenyi Biotech). The magnetic cell sorting-selected cells were analyzed by flow cytometry before and after sorting to determine the percentage of CD34+ cells, which was routinely ≥95% using this technique.

Isolation of CD34+ Cells.

CD34+ hematopoietic progenitors were isolated from the marrow of healthy donors after obtaining Institutional Review Board informed consent. Mononuclear cells were prepared using Ficoll-Hypaque and then CD34+ cells isolated to >98% purity using fluorescence activated cell sorting. This was accomplished using a FACSVantage equipped with a TurboSort option (Becton Dickinson Immunocytometry Systems, San Jose, CA). Recoveries of 50–70% and routine purities of >98% were obtained (43) . For CD34 analysis and sorting, cells were stained with directly conjugated fluorescent anti-CD34 antibody (clone 8G12, i.e., HPCA-2; Becton Dickinson Immunocytometry Systems). Dead and dying cells were excluded on the basis of 7-amino-actinomycin D staining as described previously (43) . Fluorochrome-conjugated anti-CD34 was excited at 488 nm, and fluorescence emission was detected using standard filters. CellQuest (Becton Dickinson Immunocytometry Systems) and FlowJo (TreeStar, Inc., Palo Alto, CA) software were used for data acquisition and storage.

Cells and Transfection of the bcr-abl Gene.

Human acute myelogenous leukemia HL-60/Bcr-Abl (p210) wt and HL-60/Bcr-Abl (p210) T315I cells were created by transfection of the bcr-abl gene encoding the wt p210 Bcr-Abl or the mutant bcr-abl gene encoding the Bcr-Abl (p210) T315I mutant protein, as described previously (28 , 30 , 44) . The bcr-abl cDNA was a kind gift of Dr. Thomas Skorski (Temple University, Philadelphia, PA). wt bcr-abl was cloned into the BamHI site of the plasmid vector pSTAR (44) . The p210 bcr-abl (T315I) mutant construct was created from the wt bcr-abl by site-directed mutagenesis, using Quick-change XL mutagenesis kit (Clontech, Palo Alto, CA). The resulting pSTAR bcr-Abl (wt) and p210 bcr-abl (T315I) plasmid vectors were stably transfected into HL-60 cells (44) . Stable clones were selected, subcloned by limiting dilution method, and maintained, as described previously (44) . Logarithmically growing cells were used for the studies described below.

Flow Cytometric Analysis of Cell Cycle Status and Apoptosis.

The flow cytometric evaluation of the cell cycle status and apoptosis was performed according to a previously described method (22) . The percentage of cells in the G₁, S, and G₂-M phases were calculated using Multicycle software (Phoenix Flow Systems, San Diego, CA).

Apoptosis Assessment by Annexin V Staining.

After drug treatments, cells were resuspended in 100 μl of the staining solution containing annexin V fluorescein and propidium iodide in a HEPES buffer (Annexin V-FLUOS Staining Kit; Boehringer-Mannheim and Indianapolis, IN). After incubation at room temperature for 15 min, annexin V-positive cells were estimated by flow cytometry, as described previously (42) .

Morphological Assessment of Apoptosis.

After drug treatment, 50 × 10³ cells were washed and resuspended in PBS (pH 7.3). Cytospun preparations of the cell suspensions were fixed and stained with Wright stain. Cell morphology was determined by light microscopy. In all, five different fields were randomly selected for counting of at least 500 cells. The percentage of apoptotic cells was calculated for each experiment, as described previously (22 , 42) .

Western Blot Analyses.

Western analyses of Bcr-Abl, Bcl-x_L, p-AKT, p21, p27, and acetylated histone H3 and H4 proteins and β-actin were performed using specific antisera or monoclonal antibodies (see above), as described previously (22 , 42) . Horizontal scanning densitometry was performed on Western blots by using acquisition into Adobe PhotoShop (Adobe Systems, Inc., San Jose, CA) and analysis by the NIH Image Program (NIH, Bethesda, MD). The expression of β-actin was used as a control.

Autophosphorylation of Bcr-Abl.

Cells were lysed in the lysis buffer [1% Triton X-100, 0.5% deoxycholate, 150 mm NaCl, 2.0 mm EDTA, 1.0 mm phenylmethylsulfonyl fluoride, 10 μg/ml leupeptin, 1 μg/ml Pepstin-A, 2 μg/ml Aprotinin, 25 mm NaF, 1.0 mm 4-(2-aminoethyl) benzenesulfonylfluoride hydrochloride, 20 mm P-nitrophenyl phosphate, 0.5 mm sodium orthovanadate for 30 min, and the nuclear and cellular debris cleared by centrifugation]. Protein G-agarose beads were washed twice in lysis buffer then incubated with anti-Abl monoclonal antibody [1 in 100 dilution (Santa Cruz Biotechnology] at 4°C for 2 h. After washing the protein G beads and the antibody mix with the lysis buffer, 100 μg of total cell lysates were added and incubated overnight at 4°C. The immunoprecipitates were washed three times in lysis buffer, and proteins were eluted with the SDS sample loading buffer. Proteins were separated by SDS-PAGE as described previously (22) . Immunoprecipitates were examined by Western blot analysis after transfer of proteins to nitrocellulose membranes (22 , 42) . Western blot analyses were performed with antiphospho tyrosine antibody (PharMingen, Inc.) and anti-Abl antibody to control for equal loading (42) .

Real-Time Quantitative RT-PCR.

Total RNA was isolated from cells using an RNeasy kit (Qiagen, Inc., Valencia, CA) according to the manufacturer’s instructions. Two μl of total RNA (0.1 μg) were added to 23 μl of a reaction mix (Taqman RT-PCR reagent kit; Perkin-Elmer, Boston, MA), consisting of 19.7 μl of H₂O, 0.15 μl of enzyme mix, 0.5 μl of the forward B2A2-primer, 0.5 μl of reverse B2A2-primer, 1.0 μl of bcr-abl gene-specific oligonucleotide probe, 0.5 μl of histone H1 forward primer, 0.5 μl of histone H1 reverse primer, and 0.5 μl of histone-specific oligonucleotide probe were placed in a 96-well plate. The plate was transferred to an ABI 7700 thermocycler (Perkin-Elmer), and samples were passed through 48°C, 30 min and 95°C, 10 min; these were followed by 35 amplification cycles (95°C, 15 s, 60°C, 1 min), during which fluorescent signals of bcr-abl and histone H1 probes were measured as described previously (45) . Serial dilutions of 10⁶−10¹ of bcr-abl plasmid DNA and histone plasmid DNA were used as a reference standard of fluorescent signals for copy number as described previously (45) . Histone H1 primers (forward primer, 5′-GCGAGAAATTGCTCAGGACTTTA-3′; reverse primer, 5′-GTGTCTTCAAAAAGGCCAACCA-3′) and a probe (5′-CCTCACTTGCCTCCTGCAAAGCACC-3′) were used in the reaction. p210 Bcr-Abl (e2a2) forward primer, 5′-TCCACTCAGCCACTGGATTTAA-3′; reverse primer, 5′-TGAGGCTCAAAGTCAGATGCTACT-3′, and a probe CAGAGTTCAAAAGCCCTTCAGCGGC. p185 Bcr-Abl (e1a2) forward primer, 5′-TGTGAAACTCCAGACTGTCCACA-3′; reverse primer, 5′-AAAGTCAGATGCTACTGGCCG-3′; and probe, 5′-TGACCATCAATAAGGAAGAAGCCCCTTCAGC-3′. All of the probes carried a 5′-FAM reporter label and a 3′-TAMRA quencher group. All primers and probes were synthesized by PE-Applied Biosystems (Perkin-Elmer).

Northern Blot Analysis.

Northern blot analysis of bcr-abl mRNA was performed, using a previously reported method (38) . Total RNA was extracted from treated and untreated cells using RNeasy mini kit (Qiagen, Inc.). Total RNA (20 μg) was separated on 1% agarose gel containing formaldehyde (0.22 m) and transferred overnight to an uncharged nylon membrane by capillary transfer in 10× SSC. Membranes were treated with an UV cross-linker (Stratagene) and subsequently probed with 32P-α [dCTP]-radiolabeled probe. A 413-bp fragment from the joining regions of bcr-abl generated by PCR using primers Bcr-b1 (forward) GAAGTGTTTCAGAAGCTTCTCC and ABL-A3 (reverse) GTTTGGGCTTCACACCATTCC was used as template for generating bcr-abl probe.

ChIP Assay.

ChIP analysis was performed by a slight modification of a previously described method (33) . Cells were incubated overnight at a density of 0.25 × 10⁶ cells/ml at 37°C with 5% CO₂. Next day, cells were cultured with 0, 50, 100, or 250 nm LAQ 824 for 24 h. Formaldehyde was then added to the cells to a final concentration of 1%, and the cells were gently shaken at room temperature for 10 min. After this, the cells were pelleted, suspended in 1 ml of ice-cold PBS containing protease inhibitors (Complete; Boehringer Mannheim). Cells were again pelleted, resuspended in 0.5 ml of SDS lysis buffer [1% SDS, 1.0 mm EDTA, and 50 mm Tris·HCl (pH 8.1)], and incubated on ice for 20 min. Lysates were sonicated with 15-s bursts. Debris was removed from samples by centrifugation for 20 min at 15,000 × g at 4°C. An aliquot of the chromatin preparation (100 μl) was set aside and designated as Input Fraction. The supernatants were diluted 3-fold in the immunoprecipitation buffer [0.01% SDS, 1.0% Triton X-100, 1.2 mm EDTA, 16.7 mm Tris·HCl (pH 8.1), and 150 mm NaCl], and 80 μl of 50% protein A-Sepharose slurry containing 20-μg sonicated salmon sperm DNA and 1 mg/ml BSA in the TE buffer [10 mm Tris·HCl (pH 8.0) and 1 mm EDTA] were added and incubated by rocking for 2 h at 4°C. Beads were pelleted by centrifugation, and supernatants were placed in fresh tubes with 5 μg of the antiacetylated histone H3 antibody, antiacetylated histone H4 antibody, or normal rabbit serum and incubated overnight at 4°C. Protein A-Sepharose slurry (60 μl) was added, and the samples were rocked for 1 h at 4°C. Protein A complexes were centrifuged and washed three times for 5 min each with immunoprecipitation buffer and twice for 5 min each with immunoprecipitation buffer containing 500 mm NaCl. Immune complexes were eluted twice with 250 μl of elution buffer (1% SDS, 0.1 m NaHCO₃) for 15 min at room temperature. Twenty μl of 5 m NaCl were added to the combined eluates, and the samples were incubated at 65°C for 24 h. EDTA, Tris·HCl (pH 6.5), and proteinase K were then added to the samples at a final concentration of 10 mm, 40 mm, and 0.04 μg/μl, respectively. The samples were incubated at 37°C for 30 min. Immunoprecipitated DNA (both immunoprecipitation samples and Input) was recovered by phenol/chloroform extraction and ethanol precipitation and analyzed by PCR. Actin, p21^WAF1, or bcr-specific primers were used to perform PCR on DNA isolated from ChIP experiments and Input samples. The optimal reaction conditions for PCR were determined for each primer pair. Parameters were denaturation at 95°C for 1 min and annealing at 56°C for 1 min, followed by elongation at 72°C for 1 min. PCR products were analyzed by 2.5% agarose/ethidium bromide gel electrophoresis. The primer pairs used for p21^WAF1ChIP analysis were: 5′-GGTGTCTAGGTGCTCCAGGT-3′ (dp1) and 5′-TGTCTAGGTGCTCCAG-3′ (up1). The primer pairs used for Bcr ChIP analysis were: 5′-GCCGGAGGGGAGCCAAGTGTT-3′(BCRup1), 5′-TCC CGGAAC ATG AGG TAG GTG GTG-3′ (BCR up2), 5′-CTC GAG CCC CCG TCC CTG TGC CTT TTG-3′ (BCRd1). The ratio between the immunoprecipitated DNA and Input DNA and was calculated for each treatment and primer set. The fold increases after treatment with LAQ824 was calculated from the indicated ratio.

[³H]Acetyl Lysine Labeling of Hsp90 and Its Binding to ATP-Sepharose.

K562 cells were centrifuged (1500 × g) and resuspended in RPMI medium containing 10% FBS, 0.2 mCi/ml [³H]]acetic acid (specific activity: 5 Ci/mm) and/or LAQ824 and incubated for 24 h at 37°C (pH 7.3) with gentle stirring. After this, cells were chilled on ice, centrifuged (500 × g for 5 min), and washed three times in 50 ml of PBS and 10 mm sodium butyrate. Cells were then lysed in TNESV buffer (50 mm Tris, 2 mm EDTA, 100 nm NaCl, 1 nm sodium vanadate, and 1% NP40 at pH 7.5), and total cellular proteins were quantified using the BCA protein assay. Hsp90 was immunoprecipitated from 200 μg of total protein using mouse hsp90 antibody, and immunoprecipitates were run on 10% SDS PAGE gel, fixed, dried, and autoradiographed for 48–72 h to detect [³H]]acetyl lysine-labeled hsp90 protein. Alternatively, hsp90 was affinity precipitated from 200 μg of total protein using ATP-Sepharose beads, and ATP bound proteins were eluted using 25 mm HEPES at pH 7.4, 150 mm NaCl, 1 mm NADH, 1 mm NAD, 1 mm ADP, 1 mm AMP, 1 mm DTT, 60 mm MgCl₂, and 10 mm ATP at pH 7.4. Eluted proteins were used to immunoprecipitate hsp90 using mouse hsp90 antibody, and the immunoprecipitates were run on 10% SDS PAGE gel, fixed, dried, and exposed to autoradiographed for 48–72 h to detect ATP bound [³H]]acetyl lysine-labeled hsp90 protein (35) .

Preparation of Detergent-soluble and -insoluble Fractions.

After the designated drug treatments, cells were lysed with TNSEV buffer [50 mm Tris-HCl (pH 7.5), 2 mm EDTA, 100 mm NaCl, 1 mm sodium orthovanadate, 1% NP40 containing 20 μg/ml aprotonin, 20 μg/ml leupeptin, 1 mm phenylmethylsulfonyl fluoride, 25 mm NAF, and 5 mm N-ethylmaleimide; Ref. 40] . The insoluble fraction (pellet) was solubilized with SDS buffer [80 mm Tris (pH 6.8), 2% SDS, 100 mm DTT, and 10% glycerol]. Fifty μg of proteins from the NP40 soluble and insoluble fractions were separated on 7.5% SDS-polyacrylamide gel and analyzed by Western blotting (40) .

HDAC Activity Inhibition Assay.

HDAC activity assay was performed using a kit purchased from Biomol Research Labs (Plymouth, MA), as per the manufacturer’s instructions. HeLa cell nuclear extracts (500 ng) containing HDAC activity were incubated with 50 μm of the Fluar de lys peptide substrate, which contains an acetylated lysine side chain (Biomol Research Labs) and several different concentrations of LAQ824 at room temperature. Reactions were stopped after 20 min with Fluar de lys developer and the fluorescence measured (excitation at 360 nm and emission at 460 nm; Ref. 37 ). The percentage decline in the fluorescence of the peptide substrate attributable to LAQ824 treatment was determined. TSA-mediated inhibition of the HDAC activity in the HeLa cell nuclear extract was used as the positive control.

Statistical Analyses.

Data were expressed as mean ± SE. Comparisons used student’s t test or ANOVA, as appropriate. Ps of <0.05 were assigned significance.

RESULTS

LAQ824 Treatment Induces p21 and p27 but Depletes Bcr-Abl Levels, as well as Causes Cell Cycle G₁-phase Arrest and Apoptosis of CML Cells.

We first determined the effects of LAQ824 treatment on the in vitro HDAC activity in the HeLa cell nuclear extracts. Exposure to LAQ824 (5–250 nm) resulted in inhibition of HDAC activity in a dose-dependent manner. The IC₅₀ of LAQ824 for this activity was ∼25.0 nm (Fig. 1A) ⇓ . We next determined the effect of LAQ824 on histone acetylation, Bcr-Abl, p21, and p27 levels, as well as on growth arrest and apoptosis of Bcr-Abl expressing human CML-BC cells. Treatment with 100 nM LAQ824 for 24 h increased the acetylation of histone H3 in K562 and LAMA-84 cells (Fig. 1B) ⇓ . If controlled for the total amount of histone H3, treatment with 250 nm LAQ824 further slightly increased the acetylation of histone H3 in K562 and LAMA-84 cells (Fig. 1B) ⇓ . After exposure to LAQ824, a similar increase in the histone H4 acetylation was also observed (data not shown). LAQ824-mediated histone hyperacetylation was associated with induction of p21 and p27 expression in LAMA-84 but only p27 in K562 cells (Fig. 1C) ⇓ . These results are consistent with the previous studies that SAHA induces hyperacetylation of nucleosomal histones associated with p21 but not p27 gene promoter, thereby transcriptionally up-regulating p21 but increasing p27 levels by alternative nontranscriptional mechanism (33) . Up-regulation of the expression of p27 by LAQ824 or SAHA may be attributable to the modulation of the expression and/or TK activity of Bcr-Abl in K562 and LAMA-84 cells because Griffin et al. (5 , 38) have demonstrated that Bcr-Abl down-regulates p27 levels through the activity of PI3K/AKT. Despite acetylation of the nucleosomal histones by LAQ824, p21 remained repressed in K562 cells, as was also observed after treatment with SAHA (38) . The effect of LAQ824 on the cell cycle status of K562 cells is shown in Table 1 ⇓ . The results show that exposure to 100 nm LAQ824 for 24 h markedly increased the percentage of K562 cells in the G₁ phase of the cell cycle: a change from 42.3% in the untreated cells to 96.8% in the cells treated with 100 nm LAQ824. Exposure to higher concentrations of LAQ824 (250 nm) only marginally increased the percentage of G₁-phase cells to 98.0%. The cell cycle G₁-phase accumulation was also seen, but to a lesser extent, after exposure to 0.25 or 0.50 μm of imatinib or with 0.1 or 0.25 μm of the dual Src/Bcr-Abl TK inhibitor PD180970 (9 , 42) . Similar effects were observed in LAMA-84 cells (data not shown). Importantly, treatment with 100 or 250 nm of LAQ824 for 24 h markedly depleted Bcr-Abl levels in K562 and LAMA-84 cells (Fig. 1D) ⇓ . This was associated with down-regulation of Bcl-x_L and p-AKT levels and induction of PARP cleavage activity of caspase-3 (Fig. 1D) ⇓ . Exposure of K562 and LAMA-84 cells to LAQ824 also induced a dose-dependent increase in the percentage of apoptotic cell, as detected by positive staining for annexin V (Fig. 2A) ⇓ . This was also confirmed by morphological assessment of apoptosis (data not shown).

LAQ824 Depletes wt and Mutant Bcr-Abl and Induces Apoptosis of Imatinib Refractory CML Cells.

Next, we investigated whether treatment with LAQ824 would also down-regulate the levels of the Bcr-Abl that contained the kinase domain point mutation, T315I, which has been shown to confer imatinib resistance in CML-BC cells (28) . This mutation contains the single nucleotide C to T change that results in a threonine to isoleucine substitution at position 315 of Bcr-Abl. We engineered a T315I mutation into the wt p210 bcr-abl and transfected the wt and the mutant construct into the human acute myelogenous leukemia HL-60 cells through the pSTAR expression vector (44) . Subclones of the cells were isolated that demonstrated expression of either the wt (HL-60/p210 wt) or T315I mutation-containing Bcr-Abl (HL-60/p210 T315I; Fig. 2, D and E ⇓ ). Consistent with this, whereas exposure to imatinib induced apoptosis of HL-60/p210 wt cells in a dose-dependent manner, HL-60/p210 T315I cells were resistant to imatinib-induced apoptosis (Fig. 2B) ⇓ . In contrast, treatment with LAQ824 for 24 h clearly reduced the levels of the wt and T315I mutation containing Bcr-Abl in a dose-dependent manner in the HL-60/p210 wt and HL-60/p210 T315I cells, respectively (Fig. 2, D and E) ⇓ . In both transfectants, exposure to LAQ824 also induced the PARP cleavage activity of caspase-3 and increased the percentage of annexin V-positive apoptotic cells (Fig. 2, C–E) ⇓ .

LAQ824-mediated Depletion of the mRNA of bcr-abl Is Dependent on New Protein Synthesis.

Next, we determined whether LAQ824-mediated depletion of Bcr-Abl is because of repression of the bcr-abl mRNA. After treatment with LAQ824 for varying time intervals, the ratios of the mRNA transcripts of bcr-abl and histone H1 genes were determined by the real-time quantitative RT-PCR technique. LAQ824-treated K562 cells displayed a marked decline in the ratios of the mRNA transcripts of bcr-abl to histone H1 in a time-dependent manner for up to 24 h. The ratio declined from 0.49 in the untreated cells to a ratio of 0.24 at 4 h, 0.09 at 8 h, and 0.05 at 24 h of treatment with LAQ824. Northern analysis confirmed that exposure to LAQ824 for 4 h attenuated the bcr-abl message by 50% and exposure for 16 h by 80% in K562 cells (estimated by densitometry with GAPDH mRNA as the loading control; Fig. 3A ⇓ ). This occurred before an appreciable decline in the Bcr-Abl protein levels (data not shown). We next addressed the possibility that LAQ824-induced acetylation of histones or transcription factors may indirectly, through the induction of another regulatory protein, lead to repression of the bcr-abl message (34) . Fig. 3B ⇓ demonstrates that cotreatment with cycloheximide (100 μg/ml) reversed the LAQ824 (100 nm for 5 h)-mediated decline (estimated by densitometry to be 60% as compared with the untreated cells, with GAPDH message as the loading control) in the bcr-abl mRNA in K562 cells. These results indicate that LAQ824-mediated repression of the bcr-abl message requires new protein synthesis. This supports the interpretation that LAQ824 augments the levels and activity of a transcriptional repressor for bcr-abl, an outcome that is neutralized by cotreatment with cycloheximide. This conclusion was additionally supported by the results of the ChIP analyses performed on the lysates of the untreated or LAQ824-treated K562 and LAMA-84 cells to assess the association of the promoters of the bcr-abl with the acetylated histones. As shown in Fig. 3C ⇓ , exposure to 50–250 nm of LAQ824 did not affect the bcr-abl promoter associated with acetylated histones H3 and H4 in either cell types. As a control, after exposure to 50–250 nm LAQ824 for 24 h, ∼2.0-fold more p21^WAF1 promoter DNA was associated with acetylated histones compared with the same region associated with histones from cells cultured without LAQ824 (data not shown). Previous studies have clearly shown an increase in the association of p21^WAF1 promoter DNA with acetylated histones after treatment with SAHA (33) .

LAQ824 Induces Acetylation of Hsp90, Inhibits Its Chaperone Function, and Promotes the Proteasomal Degradation of Bcr-Abl.

Prompted by the studies that the HDI depsipeptide causes acetylation and inhibition of hsp90 as a chaperone protein for Her-2 kinase (35) , we determined the effects of LAQ824 on hsp90 and its chaperone function for Bcr-Abl TK in human CML-BC cells. Fig. 4A ⇓ demonstrates that treatment with LAQ824 caused marked acetylation of hsp90 in K562 cells, detected by immunoprecipitation of hsp90, followed by immunoblotting with an antibody that recognized antiacetylated lysine residues. Additionally, after LAQ824 treatment of K562 cells cultured in a medium containing [³H]acetic acid, increased amounts of the labeled hsp90 could be immunoprecipitated from the cell lysates (Fig. 4B) ⇓ . Alternatively, these lysates from the labeled K562 cells were incubated with ATP-Sepharose resin, and the labeled hsp90 was immunoprecipitated from the eluates from the resin and estimated by autoradiography. After exposure to LAQ824, a decline in the acetylated hsp90 that could bind to the ATP-Sepharose resin was observed (Fig. 4C) ⇓ . Thus, LAQ824 clearly increased the acetylation of hsp90, which was associated with a reduced binding to ATP. Similar findings were noted in LAQ824-treated LAMA-84 cells (data not shown). Next, we evaluated the impact of these modifications on the chaperone associations of hsp90. Bcr-Abl was immunoprecipitated from cell lysates of the untreated and LAQ824 treated K562 cells, and the binding of Bcr-Abl to hsp90 versus hsp70 was determined. Fig. 4D ⇓ demonstrates that LAQ824 treatment shifts the chaperone association of Bcr-Abl from hsp90 to hsp70. A similar shift in the chaperone association of c-Raf-1 and Akt was also seen in the LAQ824-treated cells (data not shown). Previous studies have shown that the shift to the unstable multichaperone complex, including hsp70, induces the client proteins Bcr-Abl, c-Raf-1, and Akt to accumulate in the detergent-insoluble cellular fraction, as well as directs them for polyubiquitination and proteasomal degradation (40 , 46) . Fig. 4E ⇓ demonstrates that treatment with LAQ824 alone depleted the levels of Bcr-Abl, c-Raf-1, and AKT in the detergent (NP40)-soluble fraction. In contrast, LAQ824 treatment increased Bcr-Abl, but not c-Raf-1 and AKT, accumulation in detergent-insoluble fraction. This has been previously shown to precede Bcr-Abl degradation by the proteasome (40) . Consistent with this, cotreatment with the proteasome inhibitor PS341 and LAQ824 increased the accumulations of Bcr-Abl in the detergent (NP40)-insoluble cellular fraction (Fig. 4E) ⇓ . This was also seen for c-Raf-1 and AKT. Taken together, these data indicate that, by inducing acetylation of hsp90, treatment with LAQ824 disrupts the stable chaperone association of Bcr-Abl with hsp90, which promotes the proteasomal degradation of Bcr-Abl. The decline in Bcr-Abl levels was not because of its processing by caspases during LAQ824-induced apoptosis because cotreatment with the caspase inhibitor z-VAD-fmk that blocked the PARP cleavage activity of the caspases did not reverse LAQ824-mediated decline in the Bcr-Abl levels in K562 cells (data not shown).

Cotreatment with LAQ824 Enhances Imatinib or PD180970-induced Apoptosis of CML Cells.

Agents that lower Bcr-Abl levels have been shown to enhance the apoptotic effects of the Bcr-Abl TK inhibitors, e.g., imatinib, against CML-BC cells (45 , 47) . We therefore determined the effects of LAQ824 and/or imatinib or the dual Src/Abl TK inhibitor PD180970, both on the levels (and TK activity) of Bcr-Abl and on apoptosis of K562 and LAMA-84 cells. Fig. 5, A and B ⇓ , demonstrate that as compared with treatment with any agent alone, combined treatment with LAQ824 and imatinib or PD180970 for 48 h induced more apoptosis of K562 cells. Similar apoptotic effects of the combinations were also observed against LAMA-84 cells (data not shown). As compared with LAQ824 alone, treatment with 100 nm LAQ824 plus 0.25 μm imatinib or PD180970 for 24 h was associated with a greater decline (estimated by densitometry) in the levels of Bcr-Abl (60 versus 40% with LAQ824 alone; Fig. 5, C and E ⇓ ). Concomitantly, a greater decline in the autotyrosine phosphorylation of Bcr-Abl was also observed after treatment with LAQ824 and imatinib (Fig. 5D) ⇓ . Combined treatments with LAQ824 plus imatinib or PD180970, as compared with the treatment with LAQ824 alone, also produced more decline in Bcr-Abl (70 versus 50% with LAQ824 alone) and Bcl-x_L (60 versus 40% with LAQ824 alone) as well as larger increase in the expression of p27 (250 versus 80% with LAQ824) in K562 cells (estimated by densitometry) and LAMA-84 cells (data not shown). Shorter exposure intervals to LAQ824 and PD180970 or imatinib induced less effect on Bcr-Abl levels and apoptosis of K562 and LAMA-84 cells (data not shown).

LAQ824 Depletes Bcr-Abl and Induces Apoptosis of Imatinib-resistant and -sensitive Primary CML-BC Cells.

We next determined whether, similar to the findings in the cultured CML-BC cells, exposure to LAQ824 would also down-regulate the levels of Bcr-Abl and induce apoptosis of CML-BC cells derived from patients with imatinib-sensitive or -refractory CML-BC. For these studies, either samples of bone marrow-derived CD34+ leukemia progenitor cells or the mononuclear cells isolated from the peripheral blood of patients containing >90% circulating leukemia blasts were used. These samples showed varying levels of sensitivity to 0.25–1.0 μm imatinib. The percentage level of apoptosis (mean) ranged from 12.2 (sample no. 1) to 51.2 (sample no. 5; Table 2 ⇓ ). In the three imatinib-refractory samples (samples nos. 1–3), the effects of LAQ824 were determined on the intracellular levels of Bcr-Abl and Bcl-x_L, as well as on the acetylation of histone H3 and PARP cleavage activity of caspase-3. Exposure to 100 nm LAQ824 for 24 h induced acetylation of histone H3, as well as caused a decline in the Bcr-Abl and Bcl-x_L levels (Fig. 6A) ⇓ . This was also associated with the induction of the PARP cleavage activity of caspase-3. In all of the five samples, treatment with LAQ824 induced a dose-dependent increase in apoptosis (Table 2) ⇓ . These findings confirmed that as in the cultured CML-BC cells, LAQ824 exerts similar effects on histone acetylation and Bcr-Abl levels, as well as induces apoptosis of the primary CML-BC cells, regardless of their sensitivity to imatinib. In the imatinib-refractory CML-BC cells, because of the limited sample size, it was not feasible to determine whether the mechanism of resistance against imatinib was bcr-abl gene amplification or point mutations in the kinase domain of the Bcr-Abl (28 , 30) . LAQ824-induced apoptosis was also determined in CD34+ normal hemopoietic progenitor cells. Fig. 6B ⇓ demonstrates that exposure to 50, 100, and 250 nm of LAQ824 induced apoptosis of CD34+ cells in a dose-dependent manner. This was significantly lower than the percentage of apoptosis observed after exposure of cultured or primary CML-BC cells to similar concentrations of LAQ824 (P < 0.05; Fig. 2A ⇓ , Table 2 ⇓ and data not shown).

DISCUSSION

The present studies demonstrate that treatment with LAQ824 depletes Bcr-Abl levels by two mechanisms: (a) lowering the mRNA levels of bcr-abl; and (b) disrupting the chaperone association of hsp90 with Bcr-Abl, thereby promoting its degradation by the proteasome. LAQ824 lowered Bcr-Abl levels and, consequently, its TK activity. This correlated with reduced autotyrosine phosphorylation of Bcr-Abl and down-regulation of p-AKT and Bcl-x_L levels. Collectively, inhibition of these mechanisms that confer resistance against apoptosis was associated with increased PARP cleavage activity of caspase-3 and apoptosis of CML-BC cells (42) . Similar to what has been reported in other leukemia cell-types after treatment with SAHA, treatment with LAQ824 induced the transcription of p21 because of increased association of its promoter with acetylated histones (33) . In contrast, p27 promoter is not associated with acetylated histones, and increased p27 levels after treatment with SAHA have been shown to be attributable to posttranscriptional up-regulation of p27 expression (33) . This also makes it unlikely that the difference in transcriptional regulation of p21 versus p27 is because of the methylation status of the promoter regions of the two genes (46) . Regardless of the underlying mechanisms, the inductions of p21 and p27 levels may be responsible for why treatment with LAQ824 alone or in combination with imatinib induced the accumulation of CML-BC cells in the G₁ phase of the cell cycle (5) . Previous studies have demonstrated that agents that lower Bcr-Abl levels, e.g., arsenic trioxide, significantly enhance the antileukemia activity of imatinib against Bcr-Abl-expressing leukemia cells (47 , 48) . Our present findings are consistent with this, demonstrating that cotreatment with LAQ824 enhances imatinib-induced cell cycle arrest and apoptosis of CML-BC cells. Treatment with LAQ824 induced significantly more apoptosis of CML-BC than normal CD34+ hemopoietic progenitor cells. Because imatinib did not induce apoptosis of normal CD34+ hemopoietic progenitor cells, cotreatment with LAQ824 and imatinib was differentially more toxic against CML-BC cells. Recently, as alluded to above, PD180970 and related analogues were shown to potently inhibit not only Src but also Bcr-Abl TK and signal transducers and activators of transcription 5 signaling (41 , 42 , 49) . This resulted in down-regulation of the expression of several target genes, including Bcl-x_L, Mcl-1, and cyclin D2 (42 , 49 , 50) . Consequently, exposure to PD180970 induced apoptosis of CML-BC but not the Bcr-Abl-negative HL-60 cells (49 , 50) . Consistent with these observations, present studies demonstrate that treatment with LAQ824 and/or PD180970 induced cell cycle G₁-phase arrest and apoptosis of CML-BC cells (Table 1 ⇓ and Fig. 5B ⇓ ). Collectively, these findings support additional studies to determine the in vivo antileukemia activity of LAQ824 with imatinib or PD180970 against CML-BC cells.

The mRNA transcript levels of bcr-abl declined markedly after short exposure intervals to LAQ824, as determined by Northern and the real-time RT-PCR analyses. LAQ824-mediated repression of bcr-abl message was similar to what was seen in CML-BC cells after treatment with SAHA (38) . It is noteworthy that treatment with a different hydroxamic acid analogue, i.e., TSA, has been reported to inhibit the mRNA synthesis of another oncogene, erbB2, as detected by a genomically integrated ErbB2 promoter cell screen (51) . Although exposure to TSA had no effect on the open chromatin configuration of this gene, as monitored by DNase I hypersensitivity, TSA selectively destabilized and enhanced the decay of the mature ErbB2 transcripts (51) . In the present studies, the results of the ChIP analyses clearly demonstrated that treatment with LAQ824 did not affect the association of the bcr-abl promoter with the acetylated histones. This raises the possibility that HDIs may repress bcr-abl indirectly through the induction of other regulatory proteins. This is supported by our finding that cotreatment with cycloheximide reversed the LAQ824-mediated repression of the bcr-abl message. In addition to acetylating the core nucleosomal histones, HDIs may induce acetylation and alteration of DNA binding and activity of transcription factors, e.g., p53, E2F, and GATA-1 (31 , 32 , 52 , 53) . Because this was not examined in the present studies, whether LAQ824 treatment induces the activity of a transcriptional repressor of bcr-abl by this mechanism remains to be established. However, the decline in Bcr-Abl levels because of LAQ824 treatment did not appear to be caused by the processing of Bcr-Abl during LAQ824-induced apoptosis. Cotreatment with the multicaspase inhibitor z-VAD-fmk abrogated the PARP cleavage activity of the caspases, without affecting LAQ824-mediated decline in Bcr-Abl levels in CML-BC cells (data not shown).

Although the specific nuclear and/or cytosolic HDAC(s) inhibited by LAQ824, which may be involved in the acetylation of hsp90, were not elucidated, the present studies demonstrate that treatment with LAQ824 increased the acetylation of lysine residues on hsp90. This reduced the binding of hsp90 to ATP, which was associated with the disruption of the chaperone association of hsp90 with Bcr-Abl and c-Raf-1. Another hsp90 inhibitor geldanamycin or its less toxic analogue 17-allylamino-demethoxy geldanamycin has been shown to have similar effects (40 , 54) . Although attenuating the association of Bcr-Abl with the stable multichaperone complex, including hsp90, LAQ824 treatment increased the association of Bcr-Abl with the less stable multichaperone complex with hsp70. This directed the accumulation of Bcr-Abl in the detergent-insoluble cellular fraction promoting its degradation by the proteasome because cotreatment with the proteasome inhibitor PS341 with LAQ824 increased the levels of Bcr-Abl in the detergent-insoluble cellular fraction. A similar outcome was previously reported to occur, after cotreatment of CML-BC cells with 17-allylamino-demethoxy geldanamycin and PS341 (40 , 55) .

Several novel therapeutic agents have shown promising activity against imatinib-refractory CM-BC cells (9) . These agents are either able to target and inhibit the TK activity of the mutant or amplified Bcr-Abl, e.g., PD180970, farnesyl transferase inhibitor, and flavopiridol, and/or abrogate the alternative growth promoting and survival mechanisms, e.g. Lyn or Erk1/2 kinases (55, 56, 57, 58, 59) . As noted above, geldanamycin analogues and arsenic trioxide lower the levels of not only wt but also mutant Bcr-Abl and have been shown to exert antileukemia effects against imatinib-refractory CML-BC cells (42 , 60) . Recent studies from our laboratory have also demonstrated that the HDI SAHA can lower Bcr-Abl levels and induce apoptosis of imatinib-refractory CML-BC cells (38) . Present studies clearly demonstrate that LAQ824 is especially promising because it improves the selective cytotoxic effects of imatinib against CML-BC versus normal CD34+ hemopoietic progenitor cells. LAQ824 is also able to deplete not only wt but also mutant Bcr-Abl levels in imatinib-refractory CML-BC cells at concentrations as low as 100 nm. In addition, the ability of LAQ824 to attenuate c-Raf-1, Src, and AKT levels by promoting their degradation through the proteasome additionally contributes to its activity against imatinib-refractory CML-BC cells. Taken together, these attributes strongly support the in vivo testing of LAQ824 against imatinib-sensitive and -refractory forms of CML-BC.