938

Preclinical report

Epigenetic and molecular mechanisms underlying the antileukemic activity of the histone deacetylase inhibitor belinostat in human acute promyelocytic leukemia cells Jurate Savickiene, Grazina Treigyte, Giedre Valiuliene, Ieva Stirblyte and Ruta Navakauskiene Therapeutic strategies targeting histone deacetylase (HDAC) inhibition have become promising in many human malignancies. Belinostat (PXD101) is a hydroxamate-type HDAC inhibitor tested in phase I and II clinical trials in solid tumors and hematological cancers. However, little is known about the use of belinostat for differentiation therapy against acute myelogenous leukemia. Here, we characterize the antileukemia activity of belinostat as a single drug and in combination with all-trans-retinoic acid (RA) in promyelocytic leukemia HL-60 and NB4 cells. Belinostat exerted dose-dependent growth-inhibitory or proapoptotic effects, promoting cell cycle arrest at the G0/G1 or the S transition. Apoptosis was accompanied by activation of caspase 3, degradation of PARP-1, and cell cycle-dependent changes in the expression of survivin, cyclin E1, and cyclin A2. Belinostat induced a dosedependent reduction in the expression of EZH2 and SUZ12, HDAC-1, HDAC-2, and histone acetyltransferase PCAF (p300/CBP-associated factor). Belinostat increased acetylation of histone H4, H3 at K9 and H3 at K16 residues in a dose-dependent manner, but did not reduce

trimethylation of H3 at K27 at proapoptotic doses. Combined treatment with belinostat and RA dose dependently accelerated and reinforced granulocytic differentiation, accompanied by changes in the expression of CD11b, C/EBPa (CCAAT/enhancer binding protein-a), and C/EBPe. Our results concluded the usefulness of belinostat, as an epigenetic drug, for antileukemia and c differentiation therapy. Anti-Cancer Drugs 25:938–949  2014 Wolters Kluwer Health | Lippincott Williams & Wilkins.

Introduction

anticancer drugs. This includes carboplatin and paclitaxel, in a phase II study, in ovarian cancer or with proteosome inhibitor bortezomib in chronic lymphocytic leukemia or with 5-fluorouracil in colon cancer [9,10]. Belinostat exerted a dose-dependent antiproliferative effect on primary AML cells derived from a large group of consecutive patients with similar divergence observed in the presence of retinoic acid (RA), decitabine, or theophylline [11]. However, further investigation of belinostat is required to achieve effects in differentiation therapy, which is presently a main strategy for the treatment of acute promyelocytic leukemia (APL) using pharmacological doses of RA [12,13]. The APL leukemogenic fusion protein PML/RARa, generated by chromosomal translocation t(15;17), was found to repress the transcription of RA-responsive genes by recruiting many HDAC molecules, consequently promoting the development of leukemia [14]. RA binds to the fusion protein with a loss of HDAC activity and thus induces reinitiation of differentiation [15], with resultant effects on chromatin remodeling and activation of target genes. Furthermore, genetic events (DNA mutation, amplification, chromosomal rearrangement) and an epigenetic environment (DNA methylation, aberrant histone modifications) are

Histone deacetylase (HDAC) inhibitors are a new class of anticancer agents, being evaluated for the treatment of a variety of malignancies, including acute myeloid leukemia (AML) [1]. These epigenetic drugs mediate anticancer effects by promoting histone acetylation, creating a more open chromatin configuration and therefore activating genes associated with cell cycle arrest and tumor suppression. However, the use of HDAC inhibitors is limited by progressive constitutional symptoms observed in many patients. Belinostat (PXD101) is a novel low-molecular-weight hydroxamic acid HDAC inhibitor that exerts anticancer activity by inhibiting cell proliferation and inducing apoptosis in human tumor cell lines in vitro [2–4]. In vivo it inhibits the growth of human tumor xenografts with no apparent toxicity to the mice [2,5,6] and may represent a novel antitumor agent for clinical investigations. Currently, belinostat is under treatment evaluation in a number of phase I/II clinical trials in hematological malignancies and solid tumors [3,7,8]. The drug was well tolerated, with low toxicity to patients when used as a single agent or in combination with other active c 2014 Wolters Kluwer Health | Lippincott Williams & Wilkins 0959-4973 

Anti-Cancer Drugs 2014, 25:938–949 Keywords: acute myeloid leukemia, apoptosis, belinostat, differentiation, histone deacetylase inhibitors, histone modifications Department of Molecular Cell Biology, Institute of Biochemistry, Vilnius University, Vilnius, Lithuania Correspondence to Ruta Navakauskiene, PhD, Department of Molecular Cell Biology, Institute of Biochemistry, Vilnius University, Mokslininku St 12, LT-08662 Vilnius, Lithuania Tel: + 370 5 2729327; fax: + 370 5 2729196; e-mail: [email protected] Received 11 October 2013 Revised form accepted 6 March 2014

DOI: 10.1097/CAD.0000000000000122

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

Antileukemic effects of belinostat Savickiene et al.

important aspects in the development and maintenance of leukemia [16]. Notwithstanding the important anticancer effects of HDAC inhibitors, the mechanisms underlying their activities remain to be fully defined. It was established that hydroxamate HDAC inhibitors, such as LBH589 and LAQ824, target epigenetic mechanisms to reactivate repressed genes and exert antileukemia activity against human AML [17,18]. The clinical use of HDAC inhibitors in future AML therapy depends on improved characterization of the biological effects and epigenetic alterations, which may serve as biomarkers for predicting therapeutic efficacy. In this study, we extended the investigations using belinostat for APL cell lines, NB4 and HL-60. Belinostat showed dose-dependent antiproliferative and proapoptotic activity and was effective at nontoxic concentrations in the acceleration and enhancement of RA-induced differentiation with some selectivity for both cell lines. This occurred with the changes in the expression of cell cycle-associated, apoptosis-associated, and differentiation-associated genes and modification status of histones H4 and H3. In summary, this study supports the potential of belinostat as a candidate for targeted therapeutical approaches in APL.

Materials and methods Cells and culture conditions

The human promyelocytic leukemia NB4 and HL-60 cells (German Collection of Microorganisms and Cultures GmbH, Braunschweig, Germany) were cultured in RPMI 1640 medium supplemented with 10% fetal bovine serum, 100 U/ml penicillin, and 100 mg/ml streptomycin (Gibco, Grand Island, New York, USA) at 371C in a humidified 5% CO2 atmosphere. Cells were negative for mycoplasma infection using the MycoProbe detection kit (R&D Systems Europe Ltd, Minneapolis, Minnesota, USA). In each experiment, logarithmically growing cells were seeded at 5  105 cells/ml in 5 ml of medium. The stock solutions of belinostat (Selleck Chemicals LLC, Houston, Texas, USA) of 10 and 2.5 mmol/l in dimethyl sulfoxide and RA (Sigma, St Louis, Missouri, USA) of 2 mmol/l in ethanol were prepared and stored at – 201C and added directly to cell culture at a final concentration. In all experiments, the final concentration of dimethyl sulfoxide and ethanol (vehicle) did not exceed 0.05% and did not influence the results. In pretreatment experiments, cells were exposed to belinostat for 3 or 6 h, following which the drug was washed out, cells were resuspended in fresh media, and incubated with the differentiation inducer 1 mmol/l RA. Treated cells were cultured and harvested at the time-points indicated. Cell proliferation, differentiation, and viability assays

Cell proliferation was evaluated using the trypan blue exclusion test. Viable and dead (blue colored) cell count was determined using a hemocytometer. The growth

939

inhibition was calculated as follows: [(Cx – C0)–(Tx – T0)/ (Cx – C0)]  100, where C0, Cx, T0, and Tx represent the total number of cells/ml in untreated (C) and treated (T) cultures at days 0 and x (1 or 2), respectively. The degree of differentiation was evaluated by the ability of cells to reduce nitro blue tetrazolium (NBT) (Sigma) to insoluble blue–black formazan after stimulation with phorbol myristate acetate (Sigma). For statistical analysis, Student’s t-test was used to determine the significance of differences between groups for paired samples, and significance was set at *P value of 0.05 or less, **P value of 0.001 or less, and ***P value of 0.0001 or less. Flow cytometric analysis for the determination of cell cycle distribution, apoptosis, and CD11b surface biomarker

Control and treated cells were harvested, washed twice with PBS, and fixed in 70% ethanol overnight at – 201C. Fixed cells were washed twice with PBS and stained in PBS containing propidium iodide (PI) (50 mg/ml) and RNAse (0.2 mg/ml) for 30 min at 371C. Cell cycle analysis was carried out on a flow cytometer BD FACSCanto II (Becton and Dickinson, San Jose, California, USA) using BD FACSDiva software. Early and late apoptosis were analyzed using a flow cytometer after dual staining with FITC-labeled Annexin-V and PI (ApoFlowEx FITC Kit; Exbio, Prague, Czech Republic) in accordance with the manufacturer’s instruction. For the determination of CD11b, control and treated HL-60 cells were washed twice in PBS and exposed to phycoerythrin-labeled antiCD11b antibody (Dako Cytomation A/S, Copenhagen, Denmark) for 30 min at 41C in the absence of light. After washing twice with PBS, cells were fixed in PBS containing 2% paraformaldehyde for 30 min on ice, then resuspended in PBS, and analyzed by a flow cytometer. Proliferating cells were used as a control. Negative control was performed using an isotype-specific mouse phycoerythrin-conjugated IgG1 (Santa Cruz Biotechnology Inc., Heidelberg, Germany). Cell lysis and preparation of proteins

Cells (5  106 to 107) were harvested by centrifugation (500g, 6 min), washed twice in ice-cold PBS, and resuspended in 10 volumes of lysis solution (62.5 mmol/l Tris, pH 6.8, 100 mmol/l DTT, and 2% SDS, 10% glycerol). Benzonase (pure grade; Merck, Darmstadt, Germany) was added to yield a final concentration 2.5 U/ml. The lysates were prepared by homogenization through needle no. 21 on ice and then centrifuged at 20 000g for 10 min, 41C. The supernatants were immediately subjected to electrophoresis or frozen at – 761C. Protein concentrations were measured using a commercial RCDC Protein Assay (Bio Rad, Hercules, California, USA). For the preparation of histones, nuclei from control and treated cells were suspended in five volumes of 0.4 N H2SO4 by stirring and incubated overnight at 01C. The supernatant was collected by centrifugation at 15 000g for

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

940 Anti-Cancer Drugs 2014, Vol 25 No 8

10 min, 41C, and the sediment was extracted once again. After centrifugation, both extracts were combined and histones were precipitated by adding five volumes of ethanol at – 201C overnight. The precipitated histones were collected by centrifugation, washed four times with ethanol, and stored at – 201C until analysis. Gel electrophoresis and western blot analysis

Proteins were run on a 7–15% polyacrylamide gradient SDSPAGE gel using Tris-glycine buffer. After protein transfer to a PVDF membrane (Immobilon P; Millipore, Billerica, Massachusetts, USA), the membrane was blocked with PBS containing 5% BSA and 0.18% Tween-20, washed in PBS–Tween-20, and probed with the primary antibody in accordance with the manufacturer’s instructions. The membrane was subsequently washed four times with PBS–Tween-20 and then incubated with horseradish peroxidase (HRP)-linked secondary antibody for 1 h at room temperature. Immunoreactive bands were detected by enhanced chemiluminescence (GE Healthcare Amersham, Buckinghamshire, UK). Histones (5 mg) were dissolved in a buffer containing 0.9 mol/l acetic acid, 10% glycerol, 6.25 mol/l urea, and 5% b-mercaptoethanol, and separated on a 15% polyacrylamide gel containing 6 mol/l urea and 0.9 mol/l acetic acid [acid–urea (AU) gel] using 0.9 mol/l acetic acid as a buffer. Histone H3 was detected in AU gel additionally containing 4.2 mmol/l Triton X-100 and 0.9 mol/l acetic acid [triton–acid–urea (TAU) gel]. After electrophoresis, gels were stained with Brilliant Blue G-colloidal (Sigma) or blots were probed with the primary antibody against acetylated histones and secondary antibody as otherwise indicated. Immunoreactive bands were detected by enhanced chemiluminescence in accordance with the manufacturer’s instructions. The nonacetylated histone H1 band served as a control of histone extraction and protein loading. Antibodies

For western blotting, the following antibodies were used: monoclonal anti-GAPDH (ab8245) and anti-PCAF (#12188) (Abcam, Cambridge, UK), polyclonal antiEZH2 (#4905), anti-SUZ12 (#3737), anti-HDAC-1 (#6062), anti-HDAC-2 (# 9928) (Cell Signalling, Danvers, Massachusetts, USA), monoclonal anti-survivin (#sc-37616), polyclonal anti-p27 (#sc-527), anti-PARP-1 (#sc-8007) (Santa Cruz Biotechnology Inc.), and anti-ac H4K16 (#07-329) (Upstate Biotechnology Inc., Lake Placid, New York, USA). Polyclonal anti-me3 H3K27 (#07-449), anti-ac H3K9 (#07-352), monoclonal anticyclin E1 (#04-222), and anti-caspase 3 (#06-735) antibodies were from Millipore. Monoclonal anti-CD11b, C3bi receptor/RPE (# R 0841), and goat anti-rabbit (or anti-mouse) HRP-linked secondary antibodies were from Dako Cytomation A/S.

Quantitative real-time PCR

Total RNA was extracted by TRIzol (Invitrogen, Life Technologies, Carlsbad, California, USA) according to the manufacturer’s instructions, and then reverse transcribed into cDNA using the Maxima First Strand cDNA Synthesis Kit (Thermo Scientific, Vilnius, Lithuania). Quantitative real-time PCR (Q-PCR) was performed with Maxima SYBR Green qPCR Master Mix (Thermo Scientific) on the RotorGene 6000 system (Corbett Life Science, Sydney, Australia). The primers (50 –30 orientation) used were as follows: for CCAAT/enhancer binding protein-a (C/EBPa), forward (F) – GCTCGCCATGCCGGGAGAACT, reverse (R) – TGCAG GTGGCTGCTCATCG; for C/EBPE, F – CAGCCGAGGC AGCTACAATC, R – AGCCGGTACTCAAGGCTATCT; for HDAC-1, F – CAAGCTCCACATCAGTCCTTC, R – TGC GGCAGCATTCTAAGGTT; for HDAC-2, F – AGTCAAGG AGGCGGCAAAA, R – TGCGGCAGCATTCTAAGGTT; for GAPDH (glyceraldehyde-3-phosphate dehydrogenase), F – GGAAGTCAGTTCAGACTCCAGC, R – AGGCCTTTT GACTGTAATCACACC (as an internal control). The amount of mRNA was normalized to GAPDH. The relative gene expression was calculated using a comparative threshold cycle DDCt method.

Results Belinostat inhibits proliferation and induces NB4 and HL-60 cell apoptosis in a dose-dependent manner

The antiproliferative effect of belinostat, using different doses (0.2–0.5–1 mmol/l), was evaluated during 4 days of treatment. Belinostat inhibited NB4 and HL-60 cell proliferation in a dose-dependent and time-dependent manner, with the most potent effect occurring in NB4 cells (Fig. 1a). After 24 h of treatment, the drug at a dose of 0.5 mmol/l inhibited HL-60 cell proliferation more rapidly with a delay in NB4 cells. Growth inhibition of both cell lines increased up to 90% in the presence of 1 mmol/l belinostat for 24 or 48 h (Fig. 1b). Because of profound effects on cell growth and viability, we studied the influence of belinostat on cell cycle progression. Flow cytometric analysis of DNA content after NB4 and HL-60 cell treatment with belinostat for 24 h indicated dose-dependent changes (Fig. 1c). Low doses (0.2–0.5 mmol/ l) of belinostat resulted in a G0/G1-phase increase (by 8–12%), with a concomitant decrease in the S-phase. Higher concentrations (1–2 mmol/l) caused significant cell accumulation in the S-phase (to 56–63%) and decrease in both the G0/G1 and the G2/M-phase (to 12–16%), which was remarkable (9–3%) during a longer incubation time for 48 h (data not shown), indicating that the cells arrested in the S-phase transition. The cytotoxicity of belinostat was also dose dependent, where 0.2 mmol/l of the drug exerted minimal effect in both cell lines (Fig. 2a). Intermediate concentration (0.5 mmol/l) induced greater loss of cell viability, as shown by the increased percentage of dead cells with uptake

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

Antileukemic effects of belinostat Savickiene et al.

941

Fig. 1

HL-60

200

∗

150 ∗ ∗∗

100 50

∗∗

∗

∗∗

∗∗∗

1

∗∗∗

3

2

HL-60 (C) Bel 0.2 μmol/l Bel 0.5 μmol/l Bel 1 μmol/l

∗∗∗

∗∗∗

0 0

NB4

200 160 120 ∗ ∗ ∗∗

80 40 0 0

4

1

80 60

∗∗

40

∗∗∗ ∗∗∗

∗∗

24 h 48 h

100

∗∗

∗∗∗ ∗∗∗

2

3

4

24 h 48 h ∗∗ ∗

80 60

∗∗

∗∗∗ ∗∗∗

∗∗∗ ∗∗

40 20

∗

∗

1 μmol/l

∗

∗∗

∗

∗∗∗

∗∗

∗

G0/G1 S G2/M

∗

∗

/l μm

ol

/l 2

ol μm

ol

/l 1. 5

μm 1

ol

/l

∗

μm

ol

∗

/l

/l

/l μm

ol

ol μm 1

μm

ol

/l

/l

/l ol 0. 5

μm 0. 2

∗

μm

∗

∗

0.5 μmol/l

ol

∗

G0/G1 S G0M

80 70 60 50 40 30 20 10 0

μm

∗

∗

∗

C

∗

2

∗

1. 5

80 70 60 50 40 30 20 10 0

0.2 μmol/l

5

0

1 μmol/l

2

0.5 μmol/l

0.

0.2 μmol/l

(c) Cell cycle distribution (%)

∗∗

NB4

100

20 0

∗∗

120

C

Growth inhibition (%)

HL-60

120

∗∗

Time (days)

Time (days) (b)

NB4 (C) Bel 0.2 μmol/l Bel 0.5 μmol/l Bel 1 μmol/l

∗

∗

∗ ∗∗

0.

Viable cell number × 104 /ml

(a)

Dose-dependent antiproliferative effects of belinostat in HL-60 and NB4 cells. Cells were treated with the indicated doses of belinostat for 4 days. (a) Viable cell number was determined daily by counting in a hemocytometer after staining with 0.2% trypan blue. (b) The percentage of cell growth inhibition at 24–48 h. Results are mean±SD (n = 5). (c) Flow cytometric analysis of cell cycle distribution between G0/G1, S, and G2/M-phases (%) after treatment with the indicated doses of belinostat for 24 h. Results are mean±SD (n = 3); *P r 0.05, **P r 0.001, and ***P r 0.0001 indicate significant differences from untreated samples.

of trypan blue. The highest number of dead cells in both cell lines was observed in the presence of 1 mmol/l of the drug starting from day 2. NB4 and HL-60 cell exposure to belinostat dose dependently induced apoptosis as was determined by augmentation of apoptotic cell population in the subG1-phase, with a higher apoptosis-inducing effect at a dose of 1 mmol/l (Fig. 2b). Flow cytometric analysis, using two-color staining with Annexin-V–FITC and PI, showed a dose-dependent increase in a proportion of early apoptotic cells after a 24-h treatment with belinostat as compared with the control and in parallel with the augmentation of the subG1 peak (Fig. 2c). After cell incubation with higher concentrations of belinostat (1 and 2 mmol/l), the proportion of early (AnnexinV–FITC stained) apoptotic cells dose dependently increased (to 40.7 and 44.8% in NB4 and HL-60 cells, respectively, as compared with 5% in control) at 48 h of

treatment (Fig. 2d). Similarly, we found a dose-dependent increase in NB4 and HL-60 cell death, with a maximum detected after 72–96 h (Fig. 2a). Data obtained indicated that belinostat induced post-G1 arrest apoptosis in AML cells. Belinostat treatments with retinoic acid accelerate and enhance NB4 and HL-60 cell differentiation to granulocytes

For differentiation experiments, HL-60 cells were treated with a pharmacological dose of RA (1 mmol/l) [12] and belinostat added in a continuous or a temporal manner. Belinostat itself did not induce HL-60 and NB4 cell differentiation as was determined by the NBT assay and the expression of an early differentiation marker CD11b (data not shown). The time course of differentiation of HL-60 and NB4 cells subjected to cotreatment with RA

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

942

Anti-Cancer Drugs 2014, Vol 25 No 8

Fig. 2

(a)

HL-60 1 day

100

2 days

80

∗∗∗

3 days

60

∗∗ ∗

4 days

40

∗

20

∗∗ ∗∗

∗∗

∗∗

∗∗

3 days

60

4 days

∗∗

∗∗

∗

40

∗

20

∗

ol /l μm 2

1. 5

μm

ol /l

ol /l 1

μm

ol /l 0. 5

μm 0. 2

μm

C

μm ol /l

ol /l μm

2

1

1. 5

μm ol /l

ol /l 0. 5

μm 0. 2

μm

C

ol /l

P1

P1

5%

36.8%

44.8%

38.8%

/l μm

ol

/l 1

ol

/l ol μm 2 0.

AN/FITC (%)

μm

∗

/l

P1

2 μmol/l

∗

ol

P1

1 μmol/l

∗

∗∗

μm

P1

2 μmol/l

AN/FITC

∗∗

1

1 μmol/l

Control

∗∗

SubG1

C

P1

HL-60

NB4

NB4

24 h

/l

DNA content 48 h Izo-control

35 30 25 20 15 10 5 0

P1 P2 P3 P4

5

SubG1 29.5 %

∗

ol

P1 P4 P3

∗∗

0.

P2

SubG1 15.9 %

∗∗ ∗∗

μm

P3

∗

μm

SubG1. 11.2 %

P1 P4

∗

2

P2

SubG1 AN/FITC

0.

SubG1 5.1 %

P1 P4 P3

HL-60

C

P2

24 h

0. 5

SubG1 17.9 %

35 30 25 20 15 10 5 0

/l

SubG1 24.9 %

P1 P4 P3

ol

P2

Early apoptosis (%)

P1 P4 P2 P3

P1 P4 P2 P3

Early apoptosis (%)

SubG1 15.7 %

1 μmol/l

NB4 Cell count

∗∗

∗∗

(c) 0.5 μmol/l

0.2 μmol/l

Control

SubG1 6.5 %

Cell count

∗∗ ∗∗

∗∗

0

P1 P4 P2 P3

(d)

∗∗

∗

∗

24 h

HL-60

∗∗

∗∗

2 days

80

0

(b)

1 day

100

∗

∗

NB4

120

∗∗

∗∗

∗∗

∗

∗

∗∗

∗∗

∗∗

ol /l

Cell death (%)

120

40.7%

Belinostat induces dose-dependent cytotoxic effects in HL-60 and NB4 cells. Cells were treated with the indicated doses of belinostat for 4 days. (a) Cell death was evaluated after staining with 0.2% trypan blue and expressed as the percentage of nonviable cells from the total cell number. Results are mean±SD (n = 6). (b) Representative FACS-generated histograms of a hypodiploid peak (subG1) induced by belinostat at 24 h. (c) The percentages of apoptotic cells determined in a subG1 peak and by the Annexin-V test and flow cytometry after treatment of HL-60 and NB4 cells with intermediate doses of belinostat for 24 h. Results are mean±SD (n = 3); *P r 0.05, **P r 0.001, and ***P r 0.0001 indicate significant differences from untreated samples. (d) Flow cytometric analysis of early apoptosis by dual staining with Annexin-V-FITC and propidium iodide (PI) after treatment of NB4 cells with high doses of belinostat for 48 h. The percentages of cells positive with Annexin-V-FITC are indicated. The data represent one of three independent experiments showing similar results.

and belinostat at a concentration of 0.2 mmol/l showed greater differentiation toward granulocytes in comparison with treatment by RA alone (Fig. 3a). The extent of differentiation by combined treatment for 24–48 h was 1.5–1.8-fold higher than in RA-treated cells, with the most prominent acceleration in HL-60 cells (Fig. 3b). As compared with treatment by either agent alone, cotreatment with RA and belinostat decreased NB4 and HL-60 cell growth (data not shown) in parallel with increased cell arrest at the G0/G1-phase (Fig. 3c). Flow cytometric analysis after 48 h of combined treatment clearly indicated high expression of an early granulocytic marker CD11b (77.4 and 59.3% in HL-60 and NB4 cells, respectively; Fig. 3d).

Next, we determined the effect of belinostat alone and together with RA on the expression of transcription factors, C/EBPa and C/EBPE, involved in cell cycle progression and granulocytic differentiation [19,20]. Q-PCR analysis (Fig. 3e) showed that 0.2 mmol/l belinostat induced C/EBPa mRNA expression only in NB4 cells at 24–48 h of treatment with the decrease at 72 h. RA alone or together with belinostat significantly downregulated C/EBPa mRNA expression in both cell lines. Belinostat exerted a low effect on the expression of C/EBPE mRNA, which was strongly upregulated (to about 10-fold) at 24 h after NB4 and HL-60 cell treatment by RA alone or together with belinostat. C/EBPE expression gradually increased during 24–48 h and was maximal (15–25-fold

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

Antileukemic effects of belinostat Savickiene et al.

943

Fig. 3

∗ ∗∗

60 40

∗ ∗

∗

20

24 h 48 h 72 h 96 h

∗

∗

(b)

(d)

∗∗

∗

∗

150 Count

∗∗

NB4

48 h HL-60

100 50

P3

P3

P3

P2

P2

104

P2

104

104

Bel 0.2 μmol/l+RA NB4

RA NB4 52.7%

150

P1

100

0

46.1% P1

HL-60 150 100

50

77.4% P1

150

PE-A

P1 50 0

0

0

59.3%

100

50

PE-A

PE-A

PE-A

CD11b-PE (%)

RA +

RA

2 Be 4 h l2 4 h RA RA +B 48 el h 48 RA h RA 2 +B 4 h el 24 R h RA A 4 +B 8 h el 48 h

2.5 2 1.5 1 0.5 0

P4

P4

DNA content HL-60/NB4 control G0/G1 42.3%±4.5 G0/G1 59.4% G0/G1 62.3% G0/G1 54.0% G0/G1 57.2% 40.4%±4.2 S 20.9% S 21.7% S 27.2% S 29.2% S G2/M 17.3%±1.9 G2/M 19.7% G2/M 16.0% G2/M 18.8% G2/M 13.6%

RA Bel+RA

HL-60

P1

P4

P1

P2

Count

RA Bel+RA

NB4 Bel 0.2 μmol/l+RA

RA P1

P1 P4 P3

104

0

NBT+ (ratio to RA)

RA

Count

∗

48 h

Cell count

NBT+ (%)

80

HL-60 Bel 0.2 μmol/l+RA

(c)

NB4

Count

HL-60

100

Cell count

(a)

∗ ∗∗

∗∗

∗∗ ∗∗∗ ∗∗

RA RA +B el

∗∗∗ l

RA +B el

RA

l Be

RA RA +B el

Be l

∗∗

Be

∗∗ ∗∗∗ ∗∗ ∗∗∗ ∗∗∗

0

∗∗

Be l

∗∗∗∗ ∗∗∗∗ ∗∗ ∗∗

NB4 ∗∗

∗∗∗ ∗∗ ∗∗∗ ∗∗∗ ∗∗

RA

0.5

∗

24 h 48 h 72 h

RA +

∗ ∗

30 25 20 15 10 5 0

l

∗

1.5 1

24 h 48 h 72 h

∗∗

2

HL-60

NB4

Be

HL-60

2.5

C/EBPε relative expression

C/EBPα relative expression

(e)

Belinostat enhances granulocytic differentiation upon continuous treatment with retinoic acid (RA). HL-60 and NB4 cells were treated continuously with 1 mmol/l RA and 0.2 mmol/l belinostat for the indicated times. (a) Granulocytic differentiation is expressed as the percentage of nitro blue tetrazolium (NBT)-positive cells of viable cell number or (b) relative to the RA-treated control at the indicated times. Results are mean±SD (n = 3); *P r 0.05, **P r 0.001 indicate significant differences from RA-treated cell samples. (c) Flow cytometric analysis of cell cycle distribution between G0/G1, S, and G2/M-phases (%) after combined treatment with 1 mmol/l RA alone and 1 mmol/l RA with 0.2 mmol/l belinostat for 48 h. The percentages (mean±SD; n = 3) of control cell cycle distribution are indicated. (d) Representative cell surface expression of CD11b determined by flow cytometry at 48 h of combined treatment. Data (c and d) represent one of three independent experiments showing similar results. (e) C/EBPa and C/EBPE gene expression relative to control in treated samples at each time-point was determined by Q-PCR using GAPDH for normalization. Results are mean±SD (n = 3); *P r 0.05, **P r 0.001, and ***P r 0.0001 indicate significant differences from untreated samples.

increase) at 72 h when cells became differentiated. However, the concomitant treatment induced lower C/EBPE expression than the exposure to RA alone in NB4 cells with the adverse effect in HL-60 cells. Taken together, the process of cell differentiation toward granulocytes was accompanied timely by the changes in the expression of differentiation-related transcription factors, C/EBPa and C/EBPE, in a cell type-dependent manner. During the experiments, using sequential treatments, increasing concentrations (1–2–3 mmol/l) of belinostat were applied 3 h before the induction of differentiation by RA. Short exposure to the drug for 3 h induced a dose-dependent increase in the extent of differentiation, with remarkable

acceleration during the first 2 days of treatment by RA, but finally the level of differentiation was comparable with the RA-treated cells (Fig. 4a). The extension of pretreatment time to 6 h using 2 mmol/l belinostat led to higher differentiation levels at the same time-points, with the enhancement observed at 72–96 h as compared with 3-h pretreatment (Fig. 4b). Acceleration and enhancement in granulocytic differentiation was more pronounced in HL-60 cells. Flow cytometric analysis carried out at 24 h showed an increased proportion (about 2–2.5-fold) of CD11b-expressing cells in the population of both HL-60 and NB4 cells pretreated with 1 mmol/l belinostat for 3 or 6 h (Fig. 4c). The results obtained clearly show that belinostat reinforces RAinduced granulocytic differentiation.

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

944 Anti-Cancer Drugs 2014, Vol 25 No 8

Fig. 4

(a)

HL-60

100

∗∗∗ ∗∗ ∗∗

NBT+ (%)

80 60

∗∗

40

∗∗∗ ∗∗ ∗

∗∗

HL-60 RA 1 μmol/l RA 2 μmol/l RA 3 μmol/l RA

60 40

∗∗

∗∗

24 120 100 80 60 40 20 0

48

72

24

96

∗

NB4 RA 1 μmol/l RA 2 μmol/l RA 3 μmol/l RA

∗∗

∗∗ ∗

∗ ∗

∗

HL-60 RA 2 μmol/l 3 h-RA 2 μmol/l 6 h-RA

48

72

72

∗

60

∗

24

48

80

∗

0

40

∗

∗

∗

NB4 RA 2 μmol/l 3 h-RA 2 μmol/l 6 h-RA

0 0

96

96

20 24

48

72

96

Time (h)

Time (h) 24 h

Cell count

(c)

∗∗

∗∗

∗

0

0

NBT+ (%)

∗

80

20

20

(b)

NB4

100

∗∗

HL-60 1 μmol/l 3 h-RA

RA

1 μmol/l 6 h-RA

31.9%

65.1%

72.4%

P1

P1

P1

NB4 1 μmol/l 3 h-RA

RA 24.6%

43.2%

P1

P1

1 μmol/l 6 h-RA 53.1% P1

0

CD11b-PE (%) Belinostat accelerates retinoic acid (RA)-mediated differentiation upon sequential treatments. HL-60 and NB4 cells were pretreated with the indicated doses of belinostat for (a) 3 h and (b) 6 h, and then the drug was washed out and cells were cultivated in the presence of 1 mmol/l RA for the next 96 h. Granulocytic differentiation was determined daily and expressed as the percentage of nitro blue tetrazolium (NBT)-positive cells of viable cell number. Results are mean±SD (n = 3); *P r 0.05, **P r 0.001, and ***P r 0.0001 indicate significant differences from RA-treated cell samples. (c) Cell surface expression of CD11b determined by flow cytometry at 24 h of sequential treatments. The data represent one of three independent experiments showing similar results.

Belinostat induces a dose-dependent and timedependent acetylation of histone H3 and H4

To test the mechanisms by which belinostat act to alter chromatin structure, two histone modifications, representing the active chromatin state and leading to opening of the chromatin structure, were investigated. That is acetylation of histone H4 and histone H3 at K9. HL-60 and NB4 cells were treated with different doses of belinostat as a single agent. Histones were examined on AU or TAU gels that led to resolution of the distinct acetyl isoforms of histones after staining with Brilliant Blue G-colloidal. As shown in Fig. 5a–c, HL-60 and NB4 cells treated with 0.5–2 mmol/l of belinostat for a short time (1–3–6 h) included highly acetylated histone H4 isoforms up to tetra-acetylated. A concentration-dependent and time-dependent increase in the levels of acetylated histone H4 and H3 in both cell lines was observed after treatments with lower (0.4–0.8 mmol/l) and

higher (1–2 mmol/l) doses of belinostat. H4 hyperacetylation was more remarkable after 3 h of treatment with all doses of belinostat in both cell lines, but tri-acetylated and tetra-acetylated forms appeared earlier (at 1 h) in HL-60 cells. In NB4 cells, hyperacetylation of histone H4 was maintained for 3–6 h, but reduced after the drug exposure for 24 h (Fig. 5d). This was also characteristic for HL-60 cells (data not shown). For a more detailed analysis of histone H3 modifications, histones were fractionated on TAU gel and stained (Fig. 5e), which led to resolution of histone H3 from histones H2A and H2B. The positions of H3 variants (H3.1, H3.2, and H3.3) were detected, using antibodies against the total histone H3 (data not shown), and immunoblot analysis was carried out for histone H3 acetylated at K9 (Fig. 5d). As was observed after a 1-h treatment of NB4 cells with 1–3 mmol/l of belinostat, histone H3K9 acetylation (including H3.1, H3.2, and H3.3 variants) increased at 1 h of treatment

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

Antileukemic effects of belinostat Savickiene et al.

945

Fig. 5

HL-60 C 0.5 1 2 0.5 1 2 0.5 1 2 (μmol/l)

H1

(b)

Histone H4 Ac4 levels

(a)

(c)

NB4 C 0.4 0.6 0.8 0.4 0.6 0.8 (μmol/l)

HL-60

200.000

H1

150.000 100.000

Ac2 Ac1 Ac0

50.000

Ac4

0

H4

H3

1 2 3 4 5 6 7 8 9 10

Ac3

C

Ac2

1h

3h

6h

H4

Ac1 Ac0

1 2 3 4 5 6 7 8 9 10 1h

3h

1h

6h

(e)

(d)

(f)

NB4 C 1 2 3 (μmol/l) C 1 2 3

NB4 C 1 2 3 1 2 3 1 2 3 1 2 (μmol/l)

3h

1 2 3

1

2 (μmol/l)

H2A H1

Ac2 Ac1 Ac0

H3.1 H3.2 H3

H2A, H2B Ac4 Ac3 Ac2 Ac1 Ac0

1h

3h

6h

H3.3 H2B 1h

H4

3h TAU EF, H3K9 Ac

6h

24 h 1h TAU EF, stain

Belinostat induces a dose-dependent and time-dependent accumulation of histone H3 and H4 acetylation in HL-60 and NB4 cells. (a) Acid–urea (AU) gel after staining with Brilliant Blue G-colloidal represents acetyl isoforms of histone H4 (Ac0–Ac4) in HL-60 cells after treatment with the indicated doses of belinostat for 1, 3, and 6 h. The histone H1 band served as a control of histone extraction and protein loading. (b) Histone H4 hyperacetylation levels in the gel determined by scanning densitometry. (c) Acetyl isoforms of histone H3 and H4 in stained AU gel after NB4 cell treatment with intermediate doses (0.4–0.8 mmol/l) of belinostat for the indicated times. (d) Western blot analysis of acetylated H4 isoforms after NB4 cell treatment with high doses (1–3 mmol/l) of belinostat for the indicated times. (e) Triton–acid–urea (TAU) gel after staining with Brilliant Blue G-colloidal. (f) Western blot analysis of the gel using antibodies against acetylated histone H3 at K9 after NB4 cell treatment with the indicated doses of belinostat for 1–6 h. The data are representative of three independent experiments showing similar results.

and was time and a dose dependent, with a more marked effect at 6 h (Fig. 5f). Similar effects on acetylation of histone H4 and H3K9 were observed in HL-60 cells (data not shown). Thus, belinostat induced lysine hyperacetylation of both histones H3 and H4, with some timerelated differences between both cell lines. Belinostat effects on the expression of cell growth and apoptosis-related proteins

To investigate the mechanism of belinostat-induced cell death, the implication of caspase-driven apoptotic events was evaluated by examining apoptosis-associated proteins by western blot analysis using the total cell lysate of belinostat-treated cells for 24 and 48 h (Fig. 6a).

Along with the augmentation of NB4 cell death, the activation of caspase 3 was observed in a concentrationdependent and time-dependent manner. Moreover, the cleavage of procaspase 3 to active caspase 3 was confirmed by PARP-1 cleavage, which was prominent at a dose of 2 mmol/l. PARP-1 precursor protein was no longer detected at a dose of 3 mmol/l at any time of treatment in parallel with a significant reduction in cyclin A2 and p27 levels. Because belinostat at a dose of 1 or 2 mmol/l caused NB4 cell arrest at the S-phase (Fig. 1c), we did not find changes in the expression of both p27 (involved in the G1-phase) and cyclin A2 (involved in the G2/M-phase) during 24 h of treatment with 1 or 2 mmol/l of belinostat. As shown in Fig. 6b, the expression of antiapoptotic

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

946 Anti-Cancer Drugs 2014, Vol 25 No 8

Fig. 6

(a)

(b)

24 h kDa C 0.2 0.5 1 2 C 0.2 0.5 1 2

kDa

C 1 2

NB4 3 1 2

Belinostat (μmol/l)

3

124 89

PARP-1 CI. PARP-1

Belinostat (μmol/l)

98

EZH2

89

SUZ12

27

p27

50

Cyclin E1

17 12

CI. Caspase 3

60 27

Cyclin A2 p27

17

H3K27me3

62

HDAC-1

10

H4K16Ac

60

HDAC-2

93

PCAF

36

GAPDH

36

GAPDH

Time

24 h

Survivin

16.5

HL-60

48 h

NB4

Death (%) 3 15 20 35 - 27 56 62

Rrelative expression

(c) 1.8 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0

HL-60 C

-2 -1 AC DAC D H H

6h

(d)

NB4 ∗∗ 24 h

HL-60 NB4 kDa C 0.5 2 0.5 2 C 0.5 2 0.5 2

∗

Belinostat (μmol/l)

62 ∗

∗ ∗∗

∗

HDAC-1

60

HDAC-2

36 -1 -2 AC DAC HD H

Time 0

GAPDH 6h

24 h

0

6h

24 h

Effects of belinostat on the expression of proteins associated with cell growth inhibition, apoptosis, and epigenetic modifications. (a, b, d) HL-60 and NB4 cells were incubated or not in the presence of increasing doses (0.2–3 mmol/l) of belinostat for the indicated times. Western blot analysis was carried out with the indicated antibodies on the total NB4 cell lysates. GAPDH served as the loading control. (a) The percentages of dead cells in cultures where cells were lysed at the end of the incubation period and subjected to western blot analysis. (c) In HL-60 and NB4 cells treated with 2 mmol/l belinostat for 6 and 24 h, histone deacetylase-1 (HDAC-1) and HDAC-2 gene expression relative to control at each time-point was determined by Q-PCR using GAPDH for normalization. Results are mean±SD (n = 3); *P r 0.05, **P r 0.001 indicate significant differences from the untreated control.

protein survivin decreased only in cells, treated with 0.2 mmol/l belinostat, but increased drastically in a dosedependent manner in both cell lines. Similarly, belinostat induced the expression of G1 cell cycle regulators, p27 and cyclin E1, when used at low doses (0.2–0.5 mmol/l) for 24 h and caused a gradual decrease in both at cytotoxic doses (1–2 mmol/l) with the onset of growth inhibition and cell cycle arrest in both HL-60 and NB4 cells. As noted above, HL-60 cells were more sensitive and responded earlier to belinostat treatments. Belinostat dose dependently depletes histone deacetylases, EZH2, and SUZ12, and causes histone H4 and H3 modifications

We next determined the effects of belinostat, as an HDAC inhibitor, on HDAC-1 and HDAC-2 gene expression by Q-PCR and protein expression by western blot analysis. Treatment of NB4 cells with high concentrations

(1–2 mmol/l) of belinostat for 24 h dose dependently reduced both HDAC-1 and HDAC-2 protein expression, with depletion of both at a dose of 3 mmol/l similar to prolonged treatment for 48 h with 2 and 3 mmol/l of the drug (Fig. 6a). The reduction in HDAC-2 mRNA expression was observed after 6 and 24 h, whereas at the protein level, it occurred at 24 h of treatment with 2 mmol/l belinostat in NB4 and HL-60 cells (Fig. 6c and d). Treatment with 0.5 or 2 mmol/l of belinostat for 6 h was not sufficient to reduce HDAC-1 protein in both cell lines, whereas an apparent decrease was observed within 24 h preferentially in HL-60 cells (Fig. 6d). HDAC-1 mRNA levels did not change in response to 2 mmol/l of belinostat in HL-60 cells at 6 or 24 h and even increased in NB4 cells at 24 h compared with the untreated control at this time-point (Fig. 6c). Next, we determined the effect of belinostat on histone modifications, responsible for the active chromatin

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

Antileukemic effects of belinostat Savickiene et al. 947

state. PCAF (p300/CBP-associated factor), a broad-action histone acetyltransferase, participates in acetylation of histones, nonhistone proteins, many components of coactivators, and other transcription activating proteins, and also in the regulation of cell cycle progression and apoptosis [21]. In our study, the decrease in the expression of PCAF protein was dose dependent and prominent at cytotoxic concentrations (Fig. 6b) concomitant with the induction of apoptosis. In HL-60 cells, acetylation of histone H4K16, which is related to transcriptional activation, decompactization of chromatin, and maintenance of a euchromatin state [22] were induced by all doses of belinostat at 24 h with the retardation in NB4 cells. As HDACs are associated with the activity of the core polycomb repressive complex 2 (PRC2) components – SUZ12 (suppressor of zeste 12) and the catalytic subunit EZH2 (PcG enhancer of zeste homolog H2) [23,24] – we analyzed its expression in HL-60 and NB4 cells treated with increasing doses (0.2–2 mmol/l) of the drug for 24 h. Belinostat depleted EZH2 and SUZ12 in a dose-dependent manner in both cell lines, but a depletion was more pronounced at higher doses (up to 1 mmol/l) (Fig. 6b). Unexpectedly, the depletion of both components was not associated with the repressive histone epigenetic mark, trimethylated H3K27. Belinostat treatment somehow led to a dose-dependent increase in this mark in both cell lines that could occur during mitotic catastrophe, caused by high doses of belinostat or the modulation of the cell cycle regulatory or survival inhibiting genes, leading to a nonactive state of chromatin in dying cells and the reverse methylation of H3K27 and DNA [25]. These molecular and epigenetic perturbations may help to find the predictive markers of the drug activity in AML.

Discussion The precise mechanisms of belinostat action against AML remained unclear, and we extended further investigations at molecular and epigenetic levels in two different cell lines. NB4 cells are a typical promyelocytic cell line with chromosomal translocation t(15;17), bearing PML/RARa fusion protein and functional p53, whereas HL-60 cells, where PML/RARa and p53 do not exist, are stopped at a less mature stage of myeloid differentiation [26,27]. Here, we found that belinostat, when used in a broad range of concentrations (0.2–3 mmol/l), dose dependently inhibited proliferation and induced apoptosis in both cell lines (Figs 1 and 2). A previous study [11] reported similar dose-dependent effects of structurally unrelated HDAC inhibitors, including belinostat, on primary human AML cells derived from patients. However, high doses of drugs were antiproliferative and proapoptotic, whereas unaltered or even increased proliferation was observed at low or intermediate doses (including 5–80 nmol/l of belinostat) for a subset of patients [11]. We found that the toxicity of high doses

of belinostat (1–3 mmol/l) was linked to cell accumulation in the S-phase and its absence in the G2/M-phase checkpoint. This may lead to apoptosis because of belinostat-induced high histone acetylation, which impairs chromosomal condensation during mitosis [25]. As is known, aberrant mitotic entry before the completion of DNA replication, when G2 arrest is abolished, can result in mitotic catastrophe, which is accompanied by mitochondrial release of proapoptotic proteins, caspase activation, and DNA degradation [25]. In our study, the effectiveness of belinostat in cell killing was accompanied by a dose-dependent activation of caspase 3 and PARP-1 cleavage, characteristic for the intrinsic apoptotic pathway (Fig. 6a). The causal effect between cell cycle arrest and apoptosis was also associated with changes in the level of p27 and cyclin E1, the G1-phase regulators, and cyclin A2, which control the G2/M transition. Survivin, which is frequently highly expressed in leukemia cells, has a partially cell cycle-dependent subcellular distribution and actually contributes toward the mitotic spindle checkpoint [25]. In accordance with this notion, we found high accumulation of survivin in belinostat-treated cells after failure to entry into the G2/M-phase (Fig. 6b). A dose-dependent induction of apoptosis paralleled a decrease in the expression of histone acetyltransferase PCAF and polycomb proteins, EZH2 and SUZ12, in both cell lines at 24 h of belinostat treatment (Fig. 6b). Our results were found to be consistent with previous studies reporting that treatment with hydroxamate HDAC inhibitors (LBH589, LAQ824) leads to disruption of the PRC2 complex by depletion of EZH2 and downregulation of SUZ12 and EED in the cultured (including HL-60) and primary human AML cells [17], expressing significant levels of EZH2 protein. This occurred in concert with the depletion of leukemia-associated oncoproteins, growth arrest, and apoptosis and correlated with decreased levels of trimethylated and dimethylated histone H3 at K27 [17,18]. By contrast, we observed that EZH2 reduction by belinostat was associated with the increase in the H3K27me3 level after treatment with proapoptotic doses of HL-60 and NB4 cells (Fig. 6b), paralleled by a marked increase in the S-phase and a marked decrease in G1 and G2/M-phase arrest (Fig. 1c). On the basis of the data of the previous report [28] that the PRC2 complex binds the H3K27me3 mark at sites of DNA replication during incorporation of newly synthesized histones into proliferating cells, we suggest the accumulation of this mark when the majority of cells are arrested in the S-phase by higher doses of belinostat. This occurred with a timely dose-dependent depletion of EZH2 in HL-60 and NB4 cells and is consistent with the finding that EZH2 inhibition may abrogate both G1 and G2/M checkpoints and promote DNA-damaging agent-induced apoptosis in both p53 wild-type and p53-deficient cancer cells [29]. As is known, EZH2 interacts with HDAC-1 and HDAC-2 and recruits DNA methyltransferases to the PRC2

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

948 Anti-Cancer Drugs 2014, Vol 25 No 8

complex, thereby promoting DNA methylation of silenced genes that have a repressive H3K27me3 mark [23]. In primary AML cells and cell lines, the distinct patterns of HDACs, including HDAC-1 and HDAC-2, are expressed, and different HDAC inhibitors selectively alter HDACs activity [30]. In our study, belinostat affected HDAC-1 and HDAC-2 at protein and/or mRNA levels in a dose-dependent and cell type-specific manner (Fig. 6a and c), and resulted in almost complete disappearance of HDAC-1 and HDAC-2 protein in dying cells (Fig. 6a). It would appear that the actions of HDAC inhibitors depend on the cell context and differ according to the differential expression profile of HDACs, which, in all probability, determine the sensitivity to various types and dosages of HDAC inhibitors [31]. Specifically, by regulating DNMT-1, EZH2, and HDACs, HDAC inhibitors regulate epigenetic mechanisms of DNA methylation and chromatin modifications. Our data show that continuous exposure to RA with a low dose of belinostat or temporal pretreatment for 3–6 h with high doses of the drug significantly accelerated/ enhanced RA-induced HL-60 and NB4 cell differentiation that had not been reported before. This occurred in concert with early changes in acetylation of histone H3K9 and hyperacetylation of histone H4 in HL-60 cells with the delayed effect in NB4 cells (Figs 5d and 6b). Indeed, the acetylation state of lysine residues regulates local and higher order chromatin structure, including acetylation of histone H4K16, which causes chromatin decondensation, the binding of chromatin-associated proteins, and a subsequent increase in transcriptional activity [22]. The findings of our experiments show that the increase of this modification (Fig. 6b) may be implicated for apoptosis by facilitating chromatin accessibility [32] in concert with several other histone modifications, causing apoptoticrelated chromatin alterations, such as chromatin condensation or internucleosomal DNA fragmentation [33]. Our findings that belinostat is a potent inducer of RAmediated differentiation towards granulocytes were confirmed by an NBT-reducing assay, expression of cell surface marker CD11b, and changes in the expression of differentiation-related transcription factors, C/EBPa and C/EBPb. In AML, C/EBPa is a tumor suppressor, which interacts with the cyclin-dependent kinases, thereby impending cell cycle progression and blocking the differentiation of granulocytic blasts [19]. In our study, only in NB4 cells, expressing abundant levels of the C/EBPa [34], belinostat caused an increase in C/EBPa mRNA expression, which decreased in both cell lines at 72 h (Fig. 3e) in parallel with cell growth inhibition. Belinostat reinforced the RA-mediated differentiation effect, which was accompanied by downregulation of C/EBPa and significant upregulation of C/EBPE at the transcriptional level in both cell lines (Fig. 3e). As is known, the increase in C/EBPE expression, which is mediated primarily through the RARa and/or the

PML/RARa pathway [35], is associated with induced cell growth arrest, expression of CD11b, and morphological differentiation [20]. Although RA induces granulocytic differentiation in both cell lines, belinostat exerted more rapid effects in HL-60 cell differentiation (Fig. 3b) and histone H4 hyperacetylation (Fig. 5a). In addition, in HL60 cells, bearing amplified c-Myc [36], downregulation of c-Myc by RA or HDAC inhibitors is a critical event for the commitment to granulocytic differentiation [37,38]. Moreover, this occurs in association with the downregulation of cyclin E1 and upregulation of p27 [38]. Importantly, the delayed differentiation effect in NB4 cells by belinostat is consistent with our previous reports, using different HDAC inhibitors, such as sodium butyrate, FK228, and BML-210 [39–41]. Taken together, our findings indicate that belinostat induces multiple perturbations in promyelocytic leukemia cell survival, cell cycle regulatory, and differentiation mechanisms in concert with the changes in the epigenetic environment. These mechanistic insights support the rationale for a therapeutic differentiation strategy in APL that target the epigenetic regulation.

Acknowledgements This research was supported by the Research Council of Lithuania (project No. LIG-06/2012). Conflicts of interest

There are no conflicts of interest.

References 1

2

3

4

5

6

7

8

9

Bruserad O, Stapnes C, Tronstad KJ, Ryningen A, Anensen N, Gjertsed BT. Protein lysine acetylation in normal and leukemic haematopoiesis: HDACs as possible therapeutic targets in adult AML. Expert Opin Ther Targets 2006; 10:51–68. Buckley MT, Yoon J, Yee H, Chiriboga L, Libes L, Ara G, et al. The histone deacetylase inhibitor belinostat (PXD101) suppresses bladder cancer cell growth in vitro and in vivo. J Transl Med 2007; 5:49–61. Qian X, LaRochelle WJ, Ara G, Wu F, Petersen KD, Thougaard A, et al. Activity of PXD101, a histone deacetylase inhibitor, in preclinical ovarian cancer studies. Mol Cancer Ther 2006; 5:2086–2095. Ma BB, Sung F, Tao Q, Poon FF, Lui VW, Yeo W. The preclinical activity of the histone deacetylase inhibitor PXD101 (belinostat) in hepatocellular carcinoma cell lines. Invest New Drugs 2010; 28:107–114. Plumb JA, Finn PW, Williams RJ, Bandara J, Romero MR, Watkins CJ, et al. Pharmacodynamic response and inhibition of growth of human tumor xenografts by the novel histone deacetylase inhibitor PXD101. Mol Cancer Ther 2003; 29:721–728. Monks A, Hose CD, Pezzoli P, Kondapaka S, Vansant G, Petersen KD, et al. Gene expression–signature of belinostat in cell lines is specific for histone deacetylase inhibitor treatment, with a corresponding signature in xenografts. Anticancer Drugs 2009; 20:682–692. Gimsing P, Hansen M, Knudsen LM, Knoblauch P, Christensen IJ, Ooi CE, et al. A phase I clinical trial of the histone deacetylase inhibitor belinostat in patients with advanced hematological neoplasia. Eur J Haematol 2008; 81:170–176. Steele NL, Plumb JA, Vidal L, Tiornelund J, Knoblauch P, Rasmussen A, et al. A phase 1 pharmacokinetic and pharmacodynamic study of the histone deacetylase inhibitor belinostat in patients with advanced solid tumors. Clin Cancer Res 2008; 14:804–810. Tumber A, Collins LS, Petersen KD, Thougaard A, Christiansen SJ, Dejligbjerg M, et al. The histone deacetylase inhibitor PXD101 synergizes with 5-fluorouracil to inhibit colon cancer cell growth in vitro and in vivo. Cancer Chemother Pharmacol 2007; 60:275–283.

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.

Antileukemic effects of belinostat Savickiene et al. 949

10

11

12

13

14

15

16

17

18

19

20

21

22

23 24 25

26

Dizon DS, Damstrup I, Finkler NJ, Lassen U, Celano P, Glaspool R, et al. Phase II activity of belinostat (PXD-101), carboplatin, and paclitaxel in women with previously treated ovarian cancer. Int J Gynecol Cancer 2012; 22:979–986. Stapnes C, Ryningen A, Hatfield K, Øyan AM, Eide GE, Corbascio M, et al. Functional characteristics and gene expression profiles of primary acute myeloid leukemia cells identify patient subgroups that differ in susceptibility to histone deacetylase inhibitors. Int J Oncol 2007; 31:1529–1538. Breitman TR, Collins SJ, Keene BR. Terminal differentiation of human promyelocytic leukemic cells in primary culture in response to retinoic acid. Blood 1981; 57:1000–1004. Warrell RP Jr, Frankel SR, Miller WH Jr, Scheinberg DA, Itri LM, Hittelman WN, et al. Differentiation therapy of acute promyelocytic leukemia with tretinoin (all-trans-retinoic acid). N Engl J Med 1991; 324:1385–1393. Grignani F, De Matteis S, Nervi C, Tomassoni L, Gelmetti V, Cioce M, et al. Fusion partners of the retinoic acid receptor-alpha recruit histone deacetylase in promyelocytic leukemia. Nature 1998; 391:815–818. Lin RJ, Nagy L, Inoque S, Shao W, Miller WH Jr, Evans RM. Role of histone deacetylase complex in acute promyelocytic leukemia. Nature 1998; 391:811–814. Ntziachristos P, Mullenders J, Trimarchi T, Aifantis I. Mechanisms of epigenetic regulation of leukemia onset and progression. Adv Immunol 2013; 117:1–38. Fiskus W, Pranpat M, Balasis M, Herger B, Rao R, Chinnaiyan A, et al. Histone deacetylase inhibitors deplete enhancer of zeste 2 and associated polycomb repressive complex 2 proteins in human acute leukemia cells. Mol Cancer Ther 2006; 5:3096–4104. Fiskus W, Wang Y, Sreekumar A, Buckley KM, Shi H, Jillella A, et al. Combined epigenetic therapy with the histone methyltransferase EZH2 inhibitor 3-deazaneoplacin A and the histone deacetylase inhibitor panobinostat against human AML cells. Blood 2009; 114:2733–2743. Friedman AD, Keefer JR, Kummalue T, Liu H, Wang QF, Cleaves R. Regulation of granulocyte and monocyte differentiation by CCAAT/enhancer binding protein a. Blood Cells Mol Dis 2003; 31:338–341. Nakajima H, Watanabe N, Shibata F, Kitamura T, Ikeda Y, Handa M. N-terminal region of CCAAT/enhancer-binding protein epsilon is critical for cell cycle arrest, apoptosis, and functional maturation during myeloid differentiation. J Biol Chem 2006; 28:14494–14502. Cohen HY, Lavu S, Bitterman KJ, Hekking B, Imahiyerobo TA, Miller C, et al. Acetylation of the C terminus of Ku70 by CBP and PCAF controls Baxmediated apoptosis. Mol Cell 2004; 13:624–638. Shogren-Knaak M, Ishii H, Sun JM, Pazin MJ, Davie JR, Peterson CL. Histone H4-K16 acetylation controls chromatin structure and protein interactions. Science 2006; 311:844–847. Sparmann A, van Lohuizen M. Polycomb silencers control cell fate, development and cancer. Nat Rev Cancer 2006; 6:846–856. Schuettengruber B, Chourrout D, Vervoort M, Leblanc B, Cavalli G. Genome regulation by polycomb and trithorax proteins. Cell 2007; 128:735–745. Castedo M, Perfettini JL, Roumier T, Andreau K, Medema R, Kroemer G. Cell death by mitotic catastrophe: a molecular definition. Oncogene 2004; 23:2825–2837. Drexler HG, Quentmeier H, MacLeod RA, Uphoff CC, Hu ZB. Leukemia cell lines: in vitro models for the study of acute promyelocytic leukemia. Leuk Res 1995; 19:681–691.

27 Barber N, Belov L, Christopheron RI. All-trans retinoic acid induces different immunophenotypic changes on human HL-60 and NB4 myeloid leukemias. Leuk Res 2008; 32:315–322. 28 Hanse KH, Bracken A, Pasini D, Dietrich N, Gehani SS, Monrad A, et al. A model for transmission of the H3K27me3 epigenetic mark. Nat Cell Biol 2008; 10:1291–1300. 29 Wu Z, Lee ST, Qiao Y, Li Z, Lee PL, Lee YJ, et al. Polycomb protein EZH2 regulates cancer cell fate decision in response to DNA damage. Cell Death Differ 2011; 18:1771–1779. 30 Bradbury CA, Khanim FL, Hayden R, Bunce CM, White DA, Drayson MT, et al. Histone deacetylases in acute myeloid leukemia show a distinct pattern of expression that changes selectively in response to deacetylase inhibitors. Leukemia 2005; 19:1751–1759. 31 Bug G, Ottmann OG. The DAC system and associations with acute leukemias and myelodysplastic syndromes. Invest New Drugs 2010; 28 (Suppl 1):S36–S49. 32 Hajji N, Wallenborg K, Vlachos P, Fu¨llgrabe J, Hermanson O, Joseph B. Opposing effects of hMOF and SIRT1 on H4 K16 acetylation and the sensitivity to the topoisomerase II inhibitor etoposide. Oncogene 2010; 29:2192–2204. 33 Fullgrabe J, Hajji N, Joseph B. Cracking the death code: apoptosis-related histone modifications. Cell Death Differ 2010; 17:1238–1243. 34 Khwaja A, Carver J, Jones HM, Paterson D, Linch DC. Expression and dynamic modulation of the human granulocyte colony-stimulating factor receptor in immature and differentiated myeloid cells. Br J Haematol 1993; 85:254–259. 35 Park DJ, Chumakov AM, Vuong PT, Chih DY, Gombart AF, Miller WH, et al. CCAAT/enhancer binding protein epsilon is a potential retinoid target gene in acute promyelocytic leukemia treatment. J Clin Invest 1999; 103:1399–1408. 36 Collins S. The HL-60 promyelocytic leukemia cell line: proliferation, differentiation and cellular oncogene expression. Blood 1987; 70:1233–1244. 37 Li H, Wu X. Histone deacetylase inhibitor, trichostatin A, activates p21WAF1/CIP1 expression through downregulation of c-myc and release of the repression of c-myc from the promoter in human cervical cancer cells. Biochem Biophys Res Commun 2004; 324:860–867. 38 Dimberg A, Bahram F, Karlberg I, Larsson LG, Nilsson K, Oberg F. Retinoic acid-induced cell cycle arrest of human myeloid cells lines is associated with sequential down-regulation of c-Myc and cyclin E and posttranscriptional up-regulation of p27 (Kip1). Blood 2002; 99:2199–2206. 39 Savickiene J, Treigyte G, Borutinskaite V, Navakauskiene R, Magnusson KE. The histone deacetylase inhibitor FK228 distinctly sensitizes the human leukemia cells to retinoic acid-induced differentiation. Ann N Y Acad Sci 2006; 1091:368–384. 40 Savickiene J, Borutinskaite V, Treigyte G, Magnusson KE, Navakauskiene R. The novel histone deacetylase inhibitor BML-210 exerts growth inhibitory, proapoptotic and differentiation stimulating effects on the human leukemia cell lines. Eur J Pharmacol 2006; 549:9–18. 41 Savickiene J, Treigyte G, Vistartaite G, Tunaitis V, Magnusson KE, Navakauskiene R. C/EBPa and PU.1 are involved in distinct differentiation responses of acute promyelocytic leukemia HL-60 and NB4 cells via chromatin remodeling. Differentiation 2011; 81:57–67.

Copyright © Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.