The Histone Deacetylase Inhibitor Sodium Butyrate Induces DNA Topoisomerase II␣ Expression and Confers Hypersensitivity to Etoposide in Human Leukemic Cell Lines1

Home , Sodium butyrate, Sodium phenylbutyrate, TOP2B

Vol. 1, 121–131, December 2001 Molecular Cancer Therapeutics 121

Ebba U. Kurz,2 Sara E. Wilson,3 Kelly B. Leader, stimulated, protein-linked DNA complexes in both cell Brante P. Sampey, William P. Allan, lines. At concentrations with minimal effects on cell 4, 5 Jack C. Yalowich, and David J. Kroll cycle kinetics (0.4 mM in HL-60; 0.5 mM in K562), NaB Department of Pharmaceutical Sciences [E. U. K., S. E. W., K. B. L., pretreatment also modestly enhanced etoposide- B. P. S., D. J. K.], Center for Pharmaceutical Biotechnology [D. J. K.], triggered apoptosis in HL-60 cells, as determined and Program in Molecular Toxicology and Environmental Health Sciences [B. P. S., D. K. J.], School of Pharmacy, University of morphologically after acridine orange/ethidium Colorado Health Sciences Center, and University of Colorado Cancer bromide staining, and substantially increased K562 Center [E. U. K., D. J. K.], Denver, Colorado 80262, and Department of Pharmacology, University of Pittsburgh School of Medicine and Cancer growth inhibition and poly(ADP-ribose)polymerase Institute, Pittsburgh, Pennsylvania 15261 [W. P. A., J. C. Y.] cleavage after etoposide exposure. Therefore, a temporal window may exist whereby a differentiating agent may sensitize experimental leukemias to a Abstract cytotoxic antitumor agent. These results indicate that The differentiating agent and histone deacetylase histone deacetylase inhibitors should be investigated inhibitor, sodium butyrate (NaB), was shown previously for etoposide sensitization of other butyrate-responsive to cause a transient, 3–17-fold induction of human DNA hematopoietic and nonhematopoietic tumor lines in topoisomerase II␣ (topo II␣) gene promoter activity and vitro and in vivo. a 2-fold increase in topo II␣ protein early in monocytic differentiation of HL-60 cells. This observation has now Introduction been extended to other short chain fatty acids and The transcriptional and posttranscriptional regulation of DNA aromatic butyrate analogues, and evidence is topo6 II␣, a nuclear enzyme essential for completion of mi- presented that human topo II␣ promoter induction tosis (1–3), has been a primary focus of our research groups correlates closely with histone H4 acetylation status. (4–8). Drugs such as the epipodophyllotoxins (etoposide and Because increased topo II␣ expression is associated teniposide), anthracyclines, ellipticines, anthracenediones, with enhanced efficacy of topo II-poisoning antitumor and aminoacridine derivatives exert their cytotoxic activity, at drugs such as etoposide, the hypothesis tested in this least in part, by trapping topo II in a covalent complex with report was whether NaB pretreatment could sensitize DNA (9). These “cleavable complexes” (10) can act as phys- HL-60 myeloid leukemia and K562 erythroleukemia ical barriers to DNA replication (11–13), cause mitotic catas- cells to etoposide-triggered DNA damage and cell trophes (14), induce recombination events (15, 16), and/or death. A 24–72 h NaB treatment (0.4–0.5 mM) induced trigger a cytotoxicity cascade culminating in apoptosis (17– topo II␣ 2–2.5-fold in both HL-60 and K562 cells and 19). Because the class of antitumor drugs converts this es- caused a dose-dependent enhancement of etoposide- sential enzyme into a lethal instrument, it follows that tumor cell populations possessing high levels of topo II␣ are more effectively killed by these agents. Received 8/13/01; revised 9/19/01; accepted 9/25/01. Two genes exist for mammalian type II DNA topoisomer- The costs of publication of this article were defrayed in part by the ases, termed ␣ and ␤, likely attributable to evolutionary gene payment of page charges. This article must therefore be hereby marked ␣ advertisement in accordance with 18 U.S.C. Section 1734 solely to indi- duplication (20, 21). topo II is often the more predominant cate this fact. cellular form (22, 23). Cleavable complexes formed with topo 1 This work was supported by Grant RPG-97-084-01-CDD from the Amer- II␣ correlate more closely with teniposide cytotoxicity in leu- ican Cancer Society, a grant from the Cancer League of Colorado (both to ␤ D. J. K.), NIH Grant CA74972 (to J. C. Y.), and an American Cancer So- kemia cells than do those formed with topo II (24), but in ciety/Brooks Trust Fellowship Award (to S. E. W.). The experiments from other systems topo II␤ may also influence drug sensitivity the University of Colorado Cancer Center Flow Cytometry Core Facility (25, 26). The ␣ form is expressed in a proliferation-dependent were supported by NIH Grant P30 CA46934 from the National Cancer Institute. manner, whereas the ␤ form is expressed independently of 2 Present address: Department of Biological Sciences, University of Cal- growth status (27–29). Therefore, rapidly growing cells that gary, Calgary, Alberta, T2N 1N4 Canada. 3 Present address: Department of Biological Sciences, Stanford Univer- sity, Stanford, CA 94305. 4 Present address: Division of General Internal Medicine, Department of Medicine, Duke University Medical Center, Duke Comprehensive Cancer 6 The abbreviations used are: topo, topoisomerase; NaB, sodium butyr-

Center, Durham, NC 27710. ate; HDAC, histone deacetylase; IC50, drug concentration that inhibits 5 To whom requests for reprints should be addressed, at Duke University growth by 50%; PARP, poly(ADP-ribose)polymerase; ATCC, American Medical Center, Box 3020, Microbiology, Durham, NC 27710. Phone: Type Culture Collection; CMV, cytomegalovirus; ␤-gal, ␤-galactosidase; (919) 668-0281; Fax: (919) 309-4042; E-mail: [email protected]. SCFA, short-chain fatty acid; TSA, trichostatin A.

contain high topo II␣ levels are effectively killed by the topo and sodium phenylbutyrate manufactured by Elan Pharma- II-directed drugs, whereas differentiated or otherwise ceuticals (San Francisco, CA) were generous gifts of Dr. growth-arrested cells usually possess diminished topo II␣ Andrew Kraft (Division of Medical Oncology, University of levels and are intrinsically resistant to these agents (9, 30, Colorado School of Medicine). Buffers and all other chemi- 31). Similarly, acquired resistance to a topo II-directed agent cals were obtained from United States Biochemical/Amer- is often attributable to suppressed topo II␣ and/or topo II␤ sham Pharmacia Biotech (Arlington Heights, IL). Chemicals expression (8, 32, 33). were of the highest purity listed by the supplier, and molec- Several transcription factors have been linked to the high, ular biology-tested reagents were obtained wherever pos- proliferation-dependent expression of topo II␣ including c- sible. Myb and B-Myb (5), NF-M (4), NF-Y (34–37), and YB-1 (38). Mammalian Cell Culture. HL-60 human promyelocytic The Sp3 transcription factor is also known to activate the cells (ATCC CCL 240) or K562 human erythroleukemia cells ␣ human topo II gene (39). Conversely, the p53 tumor sup- (ATCC CCL 243) were obtained from the ATCC (Manassas, ␣ pressor protein represses the human topo II gene through VA). Cells were maintained as suspension cultures in either the basal transcriptional machinery (40) or via an inverted RPMI 1640 (HL-60) or Iscove’s modified Minimal Essential Ϫ CCAAT box at position 68 (41). The activated Ras pathway Medium (K562; Life Technologies, Inc.) supplemented with ␣ ␣ also stimulates both topo II activity and expression. topo II 10% fetal bovine serum, 50 units/ml penicillin G, and 50 activity is enhanced via protein interactions with extracellular ␮g/ml streptomycin sulfate. All cells were maintained at 37°C ␣ signal-regulated kinase 2 (42), and topo II trans-activation in a humidified atmosphere containing 5% CO . Cells were can be driven through a Ets-like enhancer element at position 2 carried as exponentially growing cultures by propagation at Ϫ480 of the topo II␣ promoter (43). Finally, hyperthermia also 5 ϫ 105 cells/ml every 2–3 days. induces topo II␣ expression in vitro (44) as a result of in- Cellular Histone Acetylation Assay. Nuclei were iso- creased topo II␣ mRNA stability (45). lated, and histone fractions were prepared from cell nuclei as In previous studies, the serendipitous observation was described (50) with the following modifications. Pelleted nu- made that topo II␣ transcription and gene expression was clei were resuspended in 1.35 ml of sterile, ice-cold water transiently induced by the monocytic differentiating agent, containing 1 mM phenylmethylsulfonyl fluoride and 0.15 ml of NaB (6). NaB has long been recognized as an inhibitor of 4 N H SO added dropwise with swirling and histones ex- HDAC activity (46), an effect that is, in turn, associated with 2 4 tracted overnight. After centrifugation at 12,000 ϫ g for 10 selective gene activation or repression (47–49). It was hy- min, histone supernatants were transferred to 15 ml of Fal- pothesized that as a result of increased topo II␣ expression, con polypropylene tubes and filled with cold acetone/HCl NaB-pretreated leukemia cells would exhibit increased sen- Ϫ sitivity to etoposide because of increased topo II-mediated (99:1 acetone:5 N HCl) and incubated at 20°C for 72 h. DNA damage. During the final pelleting of the histone isolates, all samples ϫ In the present study, we show that several SCFAs and were initially centrifuged for 10 min at 2,000 g in a swing- ϳ aromatic butyrate analogues also stimulate the activity of a ing-bucket rotor. The supernatants were aspirated to 1 ml, synthetic topo II␣ promoter-reporter construct in direct rela- mixed by pipetting, then transferred to a 1.5 ml of Eppendorf ϫ tion to the magnitude of histone deacetylase inhibition. NaB, tube for centrifugation at 12,000 g for 10 min at 4°C. the most efficacious stimulator of topo II␣ promoter activity Supernatants were carefully aspirated, and pellets were ly- and histone H4 acetylation, also induces the endogenous ophilized and then resuspended in loading buffer (8 M urea, expression of pharmacologically competent topo II␣ protein 10% glycerol, 0.9 N acetic acid, 5% 2-mercaptoethanol, and in both HL-60 and K562 cells. Consequently, NaB pretreat- 0.25% methyl green) for Triton-acid-urea gel electrophore- ment sensitizes human leukemia cells to etoposide cytotox- sis. Triton-acid-urea gels containing 12% polyacrylamide (in icity measured by increased fraction of morphologically ap- 7.5 M urea, 0.37% Triton X-100, 0.9 N acetic acid, 0.125% optotic cells (HL-60), enhanced growth inhibition (K562), or ammonium persulfate, and 0.125% TEMED) were prerun for stimulation of PARP proteolytic cleavage. Our results are 6 h at 115 V until constant amperage was achieved. The wells discussed in light of other recent reports of a physical rela- were loaded with 1 M cysteamine, 7.5 M urea, 0.9 N acetic tionship between topo II␣ and HDACs and indicate that other acid and run for another hour at 115 V. Solubilized histones butyrate-responsive hematopoietic and nonhematopoietic were boiled for 5 min and then loaded and subjected to cell lines should be investigated for synergism between electrophoresis at 115 V for 15–18 h. After Coomassie Blue HDAC inhibitors and classical cytotoxic agents. staining, gels were scanned, and acetylated bands were quantified on a Bio-Rad Fluor-S MultiImager (f/11, white light Materials and Methods scan, clear filter, 3 s). The intensity of uni-, bi-, and triacety- Chemicals and Enzymes. Unless noted otherwise, all cell lated histone H4 bands (denoted 1, 2, and 3, respectively) culture reagents and molecular biology grade enzymes were were combined, and data were expressed as a ratio relative obtained from Life Technologies, Inc. (Gaithersburg, MD). All to the unacetylated form (denoted 0). SCFAs were obtained from Sigma Chemical Co./Aldrich topo II Assays. Western immunoblotting for steady-state Chemical (St. Louis, MO) as sodium salts (acetate, propio- topo II␣ protein levels from HL-60 nuclear extracts was per- nate, butyrate, and caproate) or free acids (valeric and hep- formed exactly as described by Fraser et al. (6) using a tanoic acids), the latter of which were solubilized in equimolar polyclonal antiserum generated in the Yalowich laboratory to NaOH. Solutions (in 0.9% saline) of sodium phenylacetate a COOH-terminal recombinant topo II␣ peptide (7).

Transcriptional activation of a topo II␣ promoter-luciferase Cells were examined under epi-illumination at ϫ400-ϫ800. reporter construct was assessed following cellular electro- After counting 200 cells, the number of each of four cellular poration as described previously (6) using the construct, states were scored: (a) viable cells with normal nuclei (VN; Ϫ562TOP2LUC. The promoter region of this plasmid corre- bright green chromatin with organized structure); (b) viable sponds to positions Ϫ562 to ϩ90 of the human topo II␣ cells with apoptotic nuclei (VA; bright green chromatin highly 5Ј-flanking region (51), which had been cloned upstream of condensed or fragmented); (c) nonviable cells with normal the firefly luciferase cDNA in pA3LUC. Various concentra- nuclei (NVN; bright orange chromatin with organized struc- tions of NaB or other fatty acids (dissolved in sterile water, or ture); (d) nonviable cells with apoptotic nuclei (NVA; bright 0.9% saline for the aromatic analogues) were incubated in orange chromatin highly condensed or fragmented). For the triplicate with transfected cells continuously for 24 h prior to purposes of this report, data were expressed only for the harvest and quantitation of reporter gene activities. topo II␣ percentage of apoptotic cells (VA ϩ NVA/total). promoter activity was normalized to the activity derived from For K562 cells, exponentially growing cultures were ex- an internal control plasmid, pCMV-␤gal, which encodes posed to 0 or 0.5 mM NaB for 72 h and then pelleted and Escherichia coli ␤-gal under control of the CMV immediate/ exposed to increasing concentrations of etoposide in NaB- early promoter, as described previously (5, 6). free medium for1hat37°C. Cells were pelleted and washed Etoposide-stabilized topo II-DNA cleavable complexes twice with drug-free medium and then resuspended at a cell were quantified using a modification (32) of a K-SDS precip- density of 1.2 ϫ 105 cells/ml. Cells were counted 48 h later itation assay. Briefly, exponentially growing HL-60 or K562 on a model ZBF Coulter Counter (Coulter Electronics, Hi- cells (5 ϫ 105/ml) were incubated for 24 h with [14C]leucine aleah, FL). Cell growth (beyond the starting concentration of (0.2 ␮Ci/ml; 325 mCi/mmol) and [methyl-3H]]thymidine (0.6 1.2 ϫ 105 cells/ml) in etoposide-treated versus control cells ␮Ci/ml; 6.7 Ci/mmol). In the case of HL-60 cells, the labeling was ultimately expressed as a percentage of inhibition of was accompanied by a 24-h exposure to either 0 or 0.4 mM control cell growth. The 50% growth inhibitory concentration

NaB; for K562 cells, the labeling was preceded by a 48-h (IC50) was calculated from replicate concentration-response exposure to 0 or 0.5 mM NaB and continued during the curves generated from three separate experiments. labeling period for a total of a 72 h butyrate exposure. During Cleavage of PARP as a measure of apoptotic caspase the final 30 min, cells received 0–100 ␮M etoposide (Sigma activity (53) was also assessed after etoposide exposure in Chemical Co.; dissolved in DMSO) or an equivalent amount K562 cells pretreated with NaB (0.5 mM for 72 h) by immu- of vehicle, in triplicate. Cell suspensions were then pelleted noblotting for the Mr 116,000 parent band and the Mr by centrifugation at 1000 ϫ g and lysed with 500 ␮lofa 85,000–89,000 proteolytic cleavage product. Following solution containing 2.5% (w/v) SDS, 10 mM EDTA, and 0.8 SDS-PAGE with 7.5% acrylamide and electrotransfer to Im- mg/ml salmon sperm DNA. The lysate was sheared through mobilon-P membranes, blots were probed with a polyclonal a 23-gauge needle 15 times and then incubated at 65°C for antiserum obtained from Santa Cruz Biotechnology Bio- 15 min. Protein and protein-associated DNA were precipi- chemical (Santa Cruz Biotechnology, CA) at a 1:1000 dilution tated by the addition of 110 ␮lof1M KCl. Precipitates were in TBS-T [25 mM Tris-HCl (pH 7.4), 150 mM NaCl, and 0.05% recovered by centrifugation at 5000 ϫ g and washed three Tween 20] with 5% nonfat dry milk. Secondary detection of times with a solution containing 10 mM Tris-HCl (pH 8.0), 100 immune complexes was accomplished by probing with a mM KCl, 1 mM EDTA, and 0.1 mg/ml salmon sperm DNA. 1:5000 dilution of goat antirabbit-horseradish peroxidase Precipitable 14C and 3H was assessed by dissolving the conjugate (Pierce, Rockford, IL), and bands were detected pellet in 500 ␮l of water followed by liquid scintillation count- by fluorescence radiography using an enhanced chemilumi- ing. Cleavable complex data were expressed as the fold- nescence substrate (NEN). increase over control in the ratio of 3H cpm/14C cpm and Statistical Analysis. Where indicated, differences from plotted relative to etoposide concentration as the mean fold- control values were assessed by one-way ANOVA using increase Ϯ SE. Dunnett’s multiple comparison post-hoc test with GraphPad Apoptosis and Cell Growth Inhibition Measurements. Prism software. Significant differences were indicated where Ͻ The effect of a 24 h of 0.4 mM NaB pretreatment on etopo- P 0.05. side-triggered apoptosis was quantified using the ethidium bromide/acridine orange fluorescence microscopy method Results of Duke and Cohen as extensively described and docu- Several SCFA Induce Both the Activity of the DNA topo mented by Dwyer-Nield et al. (52). HL-60 cells were exposed II␣ Promoter and Acetylation of Histone H4 in HL-60 to various concentrations of etoposide during the final 2–4h Cells. It has been recognized for Ͼ20 years that SCFAs of NaB. Cells were centrifuged at 350 ϫ g for 5 min and then other than NaB are capable of enhancing the acetylation resuspended at ϳ5 ϫ 105 to 5 ϫ 106 cells/ml of RPMI 1640 state of histones, especially histone H4, by inhibiting cellular plus 10% fetal bovine serum. One ␮l of a dye mixture con- histone deacetylase activity (46, 54). Having initially demon- taining of 100 ␮g/ml acridine orange (Sigma Chemical Co.) strated that NaB was an efficacious inducer of the human and 100 ␮g/ml ethidium bromide (Sigma Chemical Co.) pre- topo II␣ promoter (6), we sought to investigate whether this pared in PBS were placed in the bottom of a 12 ϫ 75-mm was a property shared by other SCFAs and, if so, whether the glass tube, followed by 25 ␮l of cell suspension. After gentle magnitude of induction correlated with HDAC inhibitory ac- agitation, 10 ␮l of the dye-cell mixture was placed on a clean tivity. HL-60 cells were transfected with Ϫ562TOP2LUC and 2 microscope slide and then covered with a 22-mm coverslip. treated for 24 h with 1 mM concentrations of each of the

Fig. 1. Effect of SCFAs and butyrate analogues on topo II␣ promoter activity in HL-60 cells. A, the topo II␣ promoter-luciferase reporter plasmid, Ϫ562TOP2LUC (20 ␮g), and an internal control plasmid, pCMV-␤gal (1 ␮g), were cotransfected into HL-60 cells by electroporation, and cells were exposed to the sodium salts of the indicated fatty acids [acetate (C2), propionate (C3), butyrate (C4), valerate (C5), caproate (C6), heptanoate (C7), phenylacetate, or phenylbutyrate] at a concentration of 1 mM for 24 h. Cells were then processed for measurement of luciferase activity, and expression was normalized to that of ␤-galactosidase activity produced from the internal control construct. Data are expressed as relative light units divided by milliunits of ␤-gal activity; bars, SE.

SCFAs with carbon chain lengths from two to seven and the aromatic butyrate analogues, phenylacetate and phenylbutyrate (as sodium salts), followed by quantitation of luciferase reporter enzyme activity. Parallel cultures, treated identically, were subjected to analysis of histone acetylation status by electrophoresis on Triton-acid-urea gels. SCFAs as short as acetate (C2) produced a significant increase in topo II␣ promoter activity, which rose substan- Fig. 2. Correlation of increased histone H4 acetylation with topo II␣ tially as chain length increased, peaking at 9.6-fold with promoter activation. A, each group of HL-60 cells was treated with each butyrate (C4), then decreasing to 4.3-fold with caproate (C6), of the fatty acids (1 mM) for 24 h as in Fig. 1. Cells were then processed and returning to control levels with the seven-carbon fatty for isolation of nuclear histones as described in “Materials and Methods,” subjected to electrophoresis on a Triton-acid-urea polyacrylamide gel, acid, heptanoate (Fig. 1). Of the aromatic butyrate ana- and stained with Coomassie Blue. Acetylation of histone H4 was quanti- logues, only phenylbutyrate substantially activated this pro- fied by scanning the intensities of the unacetylated band and the total moter construct in HL-60 cells, although phenylacetate could acetylated bands. The intensity of uni-, bi-, and triacetylated histone H4 bands (denoted 1, 2, and 3, respectively) were combined, and data were induce the promoter Ͼ2-fold at concentrations Ն4mM (data expressed as a ratio relative to the unacetylated form (0). Differences in not shown). The biphasic response seen with the SCFAs is loading and/or histone recovery was therefore normalized by expressing acetylation as the ratio of total acetylated H4:unacetylated H4. B, graphic highly reminiscent of the relationship between carbon chain representation of the relationship between histone H4 acetylation and length and histone acetylation observed previously in HeLa topo II␣ promoter activation. The data from Fig. 1 (Y axis) and A in this cells (54). We therefore investigated whether there was a figure (X axis) are plotted with each point denoted by the associated fatty acid treatment. PA, sodium phenylacetate; PB, sodium phenylbutyrate. direct correlation between the magnitude of topo II␣ promoter activation produced by the SCFAs and the increase in histone H4 acetylation status. When similarly treated HL-60 cells were subjected to his- Our previous work demonstrated that 0.4 mM NaB treat- tone acetylation analysis (Fig. 2A), it was clear that the most ment of HL-60 cells for 24 h induced topo II␣ promoter- efficacious topo II␣ promoter inducers were also the most driven reporter enzyme activity 3–17-fold but caused only a effective inhibitors of histone H4 deacetylation. This relation- ϳ2-fold increase in steady-state topo II␣ protein levels (6). A ship is demonstrated graphically in Fig. 2B. However, this similar magnitude of NaB induction of topo II␣ protein levels concordance with topo II␣ promoter activation appeared to (2.1 Ϯ 0.3-fold; mean Ϯ SE from three separate experiments)

plateau at the very high levels of histone acetylation caused is shown in Fig. 3 by the band at Mr 170,000. Using 24 h by butyrate. This observation suggests an upper limit to this pretreatment as the optimal time for observing topo II␣ in- relationship, such that above an acetylated:unacetylated hi- duction, we investigated whether greater NaB concentra- stone H4 ratio of 3, little further induction of the topo II␣ tions (1–8mM) would further induce enzyme expression. promoter can be achieved. Although these higher NaB concentrations caused a 15–40-

Fig. 3. NaB treatment of HL-60 and K562 cells increases topo II␣ protein levels. Logarithmically growing leukemia cells were exposed to NaB (0.4 mM, 24 h for HL-60; 0.5 mM, 72 h for K562) and then lysed directly in SDS loading buffer. Proteins were resolved by electrophoresis on SDS-7.5% acrylamide gels, transferred to Immobilon-P membranes, and then immu- noblotted for human topo II␣ using an antiserum raised to the COOH terminus of the enzyme. Immunoreactivity was visualized using a goat antirabbit IgG-horseradish peroxidase secondary antibody and chemilu- Fig. 4. NaB effects on etoposide-mediated, protein-linked DNA in HL-60 ␣ minescent substrate. topo II bands were quantified by scanning on a and K562 cells. A standard K-SDS precipitation assay was used to meas- Bio-Rad Fluor-S imaging system. ure etoposide-stabilized, cleavable complex formation. HL-60 cells (A; n ϭ 3/group) and K562 cells (B; n ϭ 4 with separate experiments per- formed on separate days) were either left untreated (no NaB) or exposed to 0.4 or 0.5 mM NaB, respectively, as in Fig. 3, and labeled for the final 24 h with [14C]leucine (0.2 ␮Ci/ml) and [3H]thymidine (0.6 ␮Ci/ml). Cells fold induction of reporter enzyme activity derived from the were then treated with the indicated concentrations of etoposide (or promoter construct, endogenous topo II␣ protein levels still vehicle) for 30 min and lysed, and protein-DNA complexes were precip- itated as described in “Materials and Methods.” Protein-linked DNA was increased by only 2.5–2.7-fold at 1 mM NaB and fell to under quantitated by liquid scintillation counting to generate protein-linked DNA 2-fold control levels at higher NaB concentrations (data not as a ratio of 3H cpm/14C cpm. The data are expressed as the mean fold-increase in this ratio relative to each DMSO vehicle control; bars, SE. shown). These results are not entirely surprising in that topo f, no NaB; Œ, 0.4 mM NaB (A) and 0.5 mM NaB (B). II␣ levels rarely vary by more than 2–3-fold in proliferating cells, and forced overexpression of the enzyme is known to cause feedback inhibition of its own transcription (55). NaB Confers Hypersensitivity to Etoposide-triggered We next sought to learn whether NaB could induce topo Apoptosis (HL-60) and Enhanced Cell Growth Inhibition ␣ II protein levels in another commonly used experimental or PARP Cleavage (K562). NaB-pretreated HL-60 cells leukemia cell line, K562. Similar to HL-60 cells, K562 cells will were investigated for their apoptotic response to etoposide differentiate in response to NaB but exhibit more delayed relative to untreated controls using the ethidium bromide/ kinetics in both differentiation as well as in response to acridine orange assay of Cohen and Duke as modified (52). cytotoxic agents (18). In K562 cells treated with NaB for 72 h, This assay facilitates the microscopic identification of viable topo II␣ expression was induced 2.5 Ϯ 0.2-fold (mean Ϯ SE and nonviable apoptotic cells and distinguishes apoptotic from five separate experiments; Fig. 3). cells from those that have undergone frank necrosis. HL-60 Increased topo II␣ Expression in HL-60 and K562 Cells cells were pretreated for 24 h with 0.4 mM NaB, or sterile Correlates with Increased Etoposide-stabilized, Protein- water, and then subjected to various concentrations of eto- linked DNA. To assess whether the topo II␣ induced by NaB poside for 2 h. Both agents were then washed out, and cells in HL-60 and K562 cells was pharmacologically functional, a were incubated in drug-free medium for 2 h before process- classic protein-linked DNA assay was used in the presence ing for staining. As shown in Fig. 5A, NaB pretreatment ␮ of etoposide as an indicator of the extent of cleavable com- sensitized cells to the apoptotic effects of 30 and 100 M plex formation in intact cells (Fig. 4). Consistent with the etoposide. This phenomenon was characterized further in a ␮ NaB-mediated increase in topo II␣ expression in HL-60 and time course experiment with 30 M etoposide (Fig. 5B). At times ranging from 4 to 24 h after drug washout, NaB pre- K562 cells (Fig. 3), NaB-pretreated HL-60 and K562 cells treatment caused a modest sensitization to etoposide. Taken both displayed a greater magnitude of protein-linked DNA as together, these data indicate that NaB induction of topo II␣ a function of etoposide concentration, as compared with led to an enhancement of the apoptotic efficacy of etoposide control cells not exposed previously to NaB (Fig. 4). There- in HL-60 cells. fore, the increased levels of topo II␣ detected in Fig. 3 rep- In contrast to HL-60 cells, the erythroleukemia line K562 resented a pharmacologically functional pool of enzyme in does not readily undergo apoptosis that can be distin- both leukemic cell lines. However, it should be noted that one guished morphologically, and its cell death kinetics are de- limitation of this assay is the inability to determine the topo II layed in relation to HL-60 cells (18, 53, 56). Therefore, the isoform increasingly trapped in complexes from NaB- effect of NaB pretreatment on K562 responsiveness to eto- pretreated cells. Although these data are consistent with an poside was tested in a cell growth inhibition assay. After a ␣ increase in topo II , conclusive demonstration of the iso- 72-h pretreatment with 0.5 mM NaB, the IC50 for a 1-h eto- form(s) involved will require immunological confirmation. poside exposure was reduced by 3-fold when compared

Fig. 7. Effect of NaB pretreatment on etoposide-stimulated PARP cleavage in K562 cells. Logarithmically growing K562 cells were exposed to 0 or 0.5 mM NaB for 72 h, followed by 30 ␮M etoposide for 16 or 24 h. Cells (1 ϫ 106 per lane) were harvested, resolved by SDS-PAGE, and immu- noblotted for PARP using a polyclonal antiserum (Santa Cruz Biotechnol- ogy). Immunoreactivity was visualized using a goat antirabbit IgG-horseradish peroxidase secondary antibody and chemiluminescent substrate.

Fig. 5. Effect of NaB pretreatment on etoposide-mediated apoptosis in HL-60 cells. A, etoposide dose-response experiment. Logarithmically Table 1 Effect of each NaB treatment on cell cycle distribution of HL- growing HL-60 cells were exposed to 0 or 0.4 mM NaB for 24 h and then 60 and K562 cells incubated with the indicated etoposide concentrations for 2 h. Cells were pelleted by centrifugation, washed twice with drug-free medium, and Logarithmically growing cultures of HL-60 or K562 cells were exposed allowed to incubate for 2 h further in drug-free medium. Apoptotic cells to 0.4 mM NaB for 24 h or 0.5 mM NaB for 72 h, respectively, or an were scored by fluorescence microscopy after staining with ethidium equivalent amount of sterile water for each control. Cells were then bromide/acridine orange. The mean percentage of apoptotic cells is plot- processed for flow cytometry, and DNA content was assessed using a Coulter EPICS flow cytometer. Percentages represent the mean value Ϯ ted against etoposide concentration; bars, SE. B, etoposide time course ϭ experiment. HL-60 cells were pretreated as described in A and then SEM (n 3). ␮ exposed to 30 M etoposide for 2 h. Cells were pelleted by centrifugation, Group G –G G –M S washed twice with drug-free medium, and allowed to incubate for 4–24 h 0 1 2 further in drug-free medium. At the indicated time points, apoptotic cells HL-60 ء were quantified as described above. , statistical difference from the Control 53.63 Ϯ 0.67 10.70 Ϯ 0.19 35.68 Ϯ 0.85 respective control (P Ͻ 0.05) using Dunnett’s multiple comparison test. Ⅺ, a 0.4 mM NaB 53.87 Ϯ 0.44 12.35 Ϯ 0.26 33.75 Ϯ 0.67 control; ‚, 0.4 mM NaB. K562 Control 42.59 Ϯ 0.55 5.91 Ϯ 0.46 51.50 Ϯ 0.98 a a 0.5 mM NaB 33.02 Ϯ 0.74 6.86 Ϯ 0.64 60.12 Ϯ 0.82 a Significant differences from the respective control were determined using Dunnett’s multiple comparison test (p Ͻ 0.05).

these etoposide treatments, as evidenced by the increased

intensity of the cleaved immunoreactive band at Mr 85,000– 89,000. Effect of NaB Pretreatment on HL-60 and K562 Cell Cycle Distribution. Although the NaB enhancement of etoposide cytotoxicity in both leukemia cell lines correlated with topo II␣ induction and increased drug-induced protein-linked DNA, the potential contribution of altered cell cycle kinetics Fig. 6. Effect of NaB pretreatment on K562 cell growth after short ex- to etoposide sensitization was also investigated. topo II- posure to etoposide. Logarithmically growing K562 cells were exposed to directed drugs are known to be most effective during S- 0 or 0.5 mM NaB for 72 h and then incubated with the indicated etoposide concentrations for 1 h. Cells were pelleted by centrifugation, washed phase but can kill cells to some extent at all phases of the cell twice with drug-free medium, and incubated 48 h further in drug-free cycle (9, 12, 57). Our earlier work demonstrated that a 16-h medium, at which time total cell number was counted. The percentage of treatment of HL-60 cells with 0.4 mM NaB could trigger a very cell growth inhibition in the presence of etoposide was calculated relative 3 to total cell number in groups not treated with etoposide. Data points are transient increase in DNA synthesis (as measured by [ H]thy- expressed as the mean percentage of growth inhibition; bars, SE. Aver- midine pulse incorporation), which returned to control levels aging results from seven separate experiments, the 50% inhibitory con- Ϯ ␮ Ϯ ␮ by 24 h (6). Consistent with this earlier finding, the data in centrations (IC50) were 8.7 1.8 M and 3.1 0.5 M for cells untreated or exposed previously to 0.5 mM NaB, respectively. Table 1 indicate that 24 h of NaB pretreatment had no substantial effect on HL-60 cell cycle distribution. In fact, the

very small but statistically significant increase in G2-M dis- with control cells (Fig. 6), consistent with the increase in tribution may actually reflect the increased need for topo II␣ etoposide-mediated DNA damage. enzyme activity in chromosomal segregation. Therefore, Given that NaB pretreatment conferred etoposide hyper- NaB-mediated etoposide sensitization of HL-60 cells did not sensitivity to K562 cells, the activation of caspase activities appear to result from altered cell cycle kinetics but was was also examined by immunoblotting for cleavage of PARP. rather more likely a result of increased cellular topo II␣ con- Very little PARP cleavage was observed after either the NaB tent. pretreatment alone or with 16 or 24 h of exposure to 30 ␮M The 72-h pretreatment of K562 cells with 0.5 mM NaB etoposide (Fig. 7). In contrast, NaB pretreatment of K562 caused a modest but significant increase in S-phase distri- cells strongly enhanced the extent of PARP cleavage after bution, accompanied by a decrease of similar magnitude in

7 the G0-G1 fraction (Table 1). Although the increased S-phase expression in HL-60 cells, suggesting that global activation fraction may contribute in small part to NaB sensitization of of histone acetylation is unlikely to account for enhanced K562 cells to etoposide, it is unlikely to account entirely for topo II␣ expression. In other systems, increasing data sug- either the 3-fold decrease in etoposide IC50 shown in Fig. 6 gest the cooperation of specific acetylase/deacetylase en- or the substantial enhancement of PARP cleavage in Fig. 7. zymes with specific transcription factors via corepressor or Taken together, these data suggest that the prototypical coactivator proteins (60). For example, in transcriptional ac- HDAC inhibitor, NaB, can produce a transient hypersensitiv- tivation the yeast GCN5 histone acetyltransferase appears to ity to the antitumor action of etoposide in experimental leu- function in a complex of proteins that bridges activators to kemia cell lines by inducing functional topo II␣ expression the basal transcriptional machinery (61). Conversely, in the with a concordant increase in etoposide-stimulated, case of transcriptional repression, heterodimers of Mad and enzyme-mediated DNA damage. Max interact with SIN3 corepressor proteins to recruit RPD3 histone deacetylase activity to specific promoters, allowing lysine tails of deacetylated histones to displace activating Discussion transcription factors (62). Precisely how histone acetylation Given that NaB caused a transient increase in HL-60 cell status affects the assembly of transcriptional activators and topo II␣ expression (6), it was reasoned that a window may repressors on the endogenous human topo II␣ gene pro- exist to maximize rationally differentiation/cytotoxic drug ef- moter remains an ongoing focus of our laboratories. ficacy. In the present report, NaB-induced topo II␣ expres- Other investigators have independently identified addi- sion correlated with enhanced etoposide-stimulated, tional mechanisms by which HDAC inhibitors may converge ␣ protein-linked DNA in two human leukemia cell lines. On the on topo II regulation. HDAC1 was shown recently to interact ␣ basis of extensive literature demonstrating a positive rela- directly with topo II (63). These investigators also demon- tionship between topo II␣ levels and cytotoxic efficacy of strated that the potent fungal natural product, TSA, could sensitize CCRF-CEM cells and a number of Chinese hamster drugs that target this enzyme, the logical assumption was ovary cell lines to etoposide (63). However, topo II␣ levels or made that increased cleavable complex formation would, in activity were not assessed after TSA treatment. Shortly turn, increase etoposide cytotoxicity in NaB-pretreated cells. thereafter, Turner’s group independently confirmed the in- Indeed, NaB conferred to HL-60 cells a modest hypersensi- teraction of topo II␣ and HDAC1, but these investigators tivity to etoposide-mediated apoptosis, although in K562 demonstrated paradoxically that TSA treatment confers re- cells there was a marked inhibition of cell growth and a sistance to etoposide-induced HL-60 cell apoptosis (64). substantial enhancement of PARP cleavage. However, this end point was only measured at a single high Other limited attempts to enhance cellular sensitivity to concentration (100 ␮M) of etoposide. The present report and topo II poisons by pharmacological modulation of topo II these two other recent studies all differ in the type and levels have met with mixed success. For example, substan- phasing of HDAC inhibitor treatment and etoposide expo- ␣ ␤ tial induction of topo II and levels was observed in HL-60 sure, making it difficult at present to generalize about the cells after 96 h of treatment with all-trans retinoic acid (58). utility of combining these agents. In related work, it should However, induction was not accompanied by increases in also be noted that TSA was shown recently to induce the etoposide-induced DNA cleavage, and in fact, etoposide- activity of the mouse topo II␣ gene promoter in mouse 3T3 induced apoptosis was reduced significantly relative to con- fibroblasts (65), but this study did not investigate TSA effects trols. In contrast, one clever approach in confluent NIH3T3 on endogenous topo II␣ levels or cellular sensitivity to topo II cells to block topo II␣ transcriptional repression by NF-Y with poisons. We have subsequently observed a 3–6-fold induc- DNA minor groove binding agents effectively induced the tion of the human topo II␣ promoter with 24-h exposure to 8 enzyme and caused a dramatic 10–20-fold reduction in the TSA at 100–300 nM in several human tumor cell lines, sug- etoposide IC50 (35). These investigators argued that manip- gesting that this effect is not limited to the murine gene. ulation of topo II levels in plateau phase cell populations is Whether TSA behaves in our system with regard to increased more likely to represent situations encountered in vivo than enzyme expression and etoposide cellular sensitivity is the with exponentially growing cultures. subject of ongoing study. In the present study, a question remains concerning the The use of differentiating agents, including NaB, in clinical precise mechanism by which HDAC inhibitors induce topo management of neoplastic disease has shown some promise II␣ expression and why there is discord between the mag- (66–68), especially when combined with cytotoxic chemo- nitude of promoter activation and steady-state enzyme lev- therapy (69). However, enthusiasm has been limited because els. Although our data support a direct relationship between of the short half-life of fatty acids and a lack of mechanistic topo II␣ promoter activation (from a transfected plasmid rationale for their combination with cytotoxic agents. A construct) and the degree of histone H4 acetylation, there are search for butyrate analogues possessing a more favorable likely to be promoter-specific consequences of HDAC inhi- pharmacokinetic profile than NaB has led to clinical trials bition that mediate the magnitude, if any, of endogenous with phenylacetate, phenylbutyrate, and tributyrin (70) as gene activation. For example, the high potency HDAC inhibitor, trapoxin, induces only 2% of genes studied (59). Ectopic expression of the histone acetyltransferase P/CAF (as a 7 D. J. Kroll, unpublished observation. mimic of NaB treatment) failed to activate topo II␣ gene 8 J. C. Yalowich and D. J. Kroll, manuscript in preparation.

well as pivalyloxymethyl butyrate (AN9). Newer fungal agents elucidated (90). One possible explanation, particularly with

(TSA, as well as trapoxin and apicidin) with nanomolar inhib- regard to vincristine, is that NaB can cause G2 arrest in itory action on HDAC may also represent useful therapeutics. human breast carcinoma cells and drive DNA endoredupli- However, most clinical work to date has used the aromatic cation, resulting in polyploidy and delayed cell death (91). butyrates. Phenylbutyrate has been used in the treatment of In summary, more detailed understanding of the biological ␤-thalassemia because of its ability to induce fetal hemoglo- consequences of altering histone acetylation may lead to bin (71) and in children with urea-cycle disorders (72) be- opportunities for synergistic combination chemotherapy because of its glutamine-scavenging activity. In fact, phenyl- tween HDAC inhibitors and classical cytotoxic agents. It butyrate was used recently to facilitate the birth of a child should also be appreciated that HDAC inhibitors possess harboring a heterozygous mutation for ornithine transcar- other nonhistone activities that may contribute to their sin- bamylase (73). gle-agent and combination antitumor effects, and moreover, With regard to cancer therapy, others have investigated some of these effects may be unique to the HDAC inhibitor HDAC inhibitors for efficacy as single-agent antitumor drugs. in question. Our current work details a direct link between This approach arose originally from the observation that a HDAC inhibitors, the induction of topo II␣, and the conse- number of nonhematopoietic cell lines (including melanoma quent enhancement of etoposide-triggered DNA damage. and carcinoma of the breast, ovary, lung, prostate, and co- However, further studies on HDAC inhibitors in post-DNA lon) were responsive to the differentiating effects of butyrate damage processing events may provide insights for maxi- or butyrate analogues (74–78). A synthetic benzamide HDAC mizing the anticancer efficacy of these agents. inhibitor (MS-27–275) also possesses antitumor activity in a variety of human tumor xenografts in nude mice (79). Simi- Acknowledgments larly, sodium phenylbutyrate can trigger apoptosis in pros- We dedicate this work to the memory of Dr. Alan P. Wolffe (1959–2001), tate carcinoma cells (80). In colon carcinoma, NaB causes whose untimely passing (reported in Science, 293: 1065, 2001) represents cell cycle arrest by inducing p21 (78), but this effect does not a great loss to the field of epigenetic transcriptional control by histone require p53 (81, 82). The latter finding is provocative in that acetylation. The foundation work of Drs. Tessa L. Brandt and David J. the HDAC inhibitors may activate normally p53-dependent Fraser are recognized in providing the hypothesis tested in this report. The technical assistance of Pegge M. Halandras and Charles T. Nuttelman are apoptotic machinery in cells lacking functional p53 (83, 84). also recognized for the generation of preliminary data leading to the Some of the cytostatic effects of phenylacetate and phenyl- current studies. We thank Dr. J. John Cohen (University of Colorado butyrate may also be mediated by their action as ligands for School of Medicine, Division of Immunology) for access to his fluores- the peroxisome proliferator-activated receptor ␥ (85). Al- cence microscope. D. J. K. also thanks Dr. Kenneth N. Kreuzer (Duke though it remains to be demonstrated whether any of these University Medical Center) and the members of his laboratory for their constructive comments and stimulating discussion of these studies. other cell types also exhibit NaB-inducible topo II␣ expression or whether peroxisome proliferator-activated receptor ␥ affects topo II␣ expression, HDAC inhibitors are already be- References ing tested in the clinic against several solid tumor types. 1. Holm, C., Goto, T., Wang, J. C., and Botstein, D. DNA topoisomerase Finally, HDAC inhibitors may have antitumor effects that may II is required at the time of mitosis in yeast. Cell, 41: 553–563, 1985. only be apparent when in vivo models are used. For example, 2. Holm, C., Stearns, T., and Botstein, D. DNA topoisomerase II must act at mitosis to prevent nondisjunction and chromosome breakage. Mol. TSA was shown recently to act as an angiogenesis inhibitor Cell. Biol., 9: 159–168, 1989. in human tumor xenografts of nude mice in part by suppress- 3. Uemura, T., Ohkura, H., Adachi, Y., Morino, K., Shiozaki, K., and ing hypoxia-responsive tumor suppressor genes (86). Yanagida, M. DNA topoisomerase II is required for condensation and Although our work has focused on the ability of HDAC separation of mitotic chromosomes in S. pombe. Cell, 50: 917–925, 1987. inhibitors to influence topo II-dependent drug action, evi- 4. Brandt, T. L., and Kroll, D. J. NF-M trans-activates the human DNA dence suggests that HDAC inhibitors may also be useful topoisomerase II␣ promoter independently of c-Myb in HL-60 cells. Leuk. combination chemotherapy agents via actions independent Res., 21: 711–720, 1997. of and/or downstream from topo II-mediated DNA damage. 5. Brandt, T. L., Fraser, D. J., Leal, S., Halandras, P. M., Nelson, A. R., and ␣ Phenylbutyrate, but not NaB, enhanced doxorubicin cyto- Kroll, D. J. c-Myb trans-activates the human DNA topoisomerase II gene promoter. J. Biol. Chem., 272: 6278–6284, 1997. toxicity in culture, presumably by a loss of numerous cellular 6. Fraser, D. J., Brandt, T. L., and Kroll, D. J. Topoisomerase II␣ promoter antioxidant defenses (87), although topo II␣ levels were not trans-activation early in monocytic differentiation of HL-60 human leuke- quantified. In another study (88), pivalyloxymethylbutyrate mia cells. Mol. Pharmacol., 47: 696–706, 1995. (AN9) was shown to enhance anthracycline efficacy but via 7. Ritke, M. K., Allan, W. P., Fattman, C., Gunduz, N. N., and Yalowich, decreased rate of anthracycline aglycone formation and a J. C. Reduced phosphorylation of topoisomerase II in etoposide-resistant reduction in NAD(P)H:quinone oxidoreductase and cyto- human leukemia K562 cells. Mol. Pharmacol., 46: 58–66, 1994. chrome P-450 reductase activities. Oxidative stress attribut- 8. Ritke, M. K., Roberts, D., Allan, W. P., Raymond, J., Bergoltz, V. V., and able to reactive oxygen species has also been implicated in Yalowich, J. C. Altered stability of etoposide-induced topoisomerase II- butyrate sensitization of colorectal carcinoma cells to tumor DNA complexes in resistant human leukaemia K562 cells. Br. J. Cancer, 69: 687–697, 1994. necrosis factor-␣ and Fas ligand (89). Reactive oxygen spe- 9. Corbett, A. H., and Osheroff, N. When good enzymes go bad: conver- cies may also contribute to p53-independent induction of sion of topoisomerase II to a cellular toxin by antineoplastic drugs. Chem. p21 during growth arrest (83). In addition, NaB potentiates Res. Toxicol., 6: 585–597, 1993. apoptosis in retinoblastoma cells treated with vincristine or 10. Nelson, E. M., Tewey, K. M., and Liu, L. F. Mechanism of antitumor cisplatin, although the mechanism of this effect was not drug action: poisoning of mammalian DNA topoisomerase II on DNA by

4Ј-(9-acridinylamino)-methanesulfon-m-anisidide. Proc. Natl. Acad. Sci. action in human, mouse, and Chinese hamster ovary cells. Cancer Res., USA, 81: 1361–1365, 1984. 47: 3973–3979, 1987. 11. Catapano, C. V., Carbone, G. M., Pisani, F., Qiu, J., and Fernandes, 29. Woessner, R. D., Mattern, M. R., Mirabelli, C. K., Johnson, R. K., and D. J. Arrest of replication fork progression at sites of topoisomerase Drake, F. H. Proliferation- and cell cycle-dependent differences in expres- II-mediated DNA cleavage in human leukemia CEM cells incubated with sion of the 170 kilodalton and 180 kilodalton forms of topoisomerase II in VM-26. Biochemistry, 36: 5739–5748, 1997. NIH-3T3 cells. Cell Growth Differ., 2: 209–214, 1991. 12. D’Arpa, P., Beardmore, C., and Liu, L. F. Involvement of nucleic acid 30. Davies, S. M., Robson, C. N., Davies, S. L., and Hickson, I. D. Nuclear synthesis in cell killing mechanisms of topoisomerase poisons. Cancer topoisomerase II levels correlate with the sensitivity of mammalian cells to Res., 50: 6919–6924, 1990. intercalating agents and epipodophyllotoxins. J. Biol. Chem., 263: 17724– 13. Hong, G., and Kreuzer, K. N. An antitumor drug-induced topoisomer- 17729, 1988. ase cleavage complex blocks a bacteriophage T4 replication fork in vivo. 31. Kaufmann, S. H., McLaughlin, S. J., Kastan, M. B., Liu, L. F., Karp, Mol. Cell. Biol., 20: 594–603, 2000. J. E., and Burke, P. J. Topoisomerase II levels during granulocytic matu- 14. Lock, R. B., and Stribinskiene, L. Dual modes of death induced by ration in vitro and in vivo. Cancer Res., 51: 3534–3543, 1991. etoposide in human epithelial tumor cells allow Bcl-2 to inhibit apoptosis 32. Ritke, M. K., and Yalowich, J. C. Altered gene expression in human without affecting clonogenic survival. Cancer Res., 56: 4006–4012, 1996. leukemia K562 cells selected for resistance to etoposide. Biochem. Phar- 15. Berger, N. A., Chatterjee, S., Schmotzer, J. A., and Helms, S. R. macol., 46: 2007–2020, 1993. Etoposide (VP-16-213)-induced gene alterations: potential contribution to 33. Webb, C. D., Latham, M. D., Lock, R. B., and Sullivan, D. M. Atten- cell death. Proc. Natl. Acad. Sci. USA, 88: 8740–8743, 1991. uated topoisomerase II content directly correlates with a low level of drug 16. Neece, S. H., Carles-Kinch, K., Tomso, D. J., and Kreuzer, K. N. Role resistance in a Chinese hamster ovary cell line. Cancer Res., 51: 6543– of recombinational repair in sensitivity to an antitumour agent that inhibits 6549, 1991. bacteriophage T4 type II DNA topoisomerase. Mol. Microbiol., 20: 1145– 34. Isaacs, R. J., Harris, A. L., and Hickson, I. D. Regulation of the human 1154, 1996. topoisomerase II␣ gene promoter in confluence-arrested cells. J. Biol. 17. Hashimoto, H., Chatterjee, S., and Berger, N. A. Inhibition of etopo- Chem., 271: 16741–16747, 1996. side (VP-16)-induced DNA recombination and mutant frequency by Bcl-2 35. Tolner, B., Hartley, J. A., and Hochhauser, D. Transcriptional regula- protein overexpression. Cancer Res., 55: 4029–4035, 1995. tion of topoisomerase II␣ at confluence and pharmacological modulation 18. Ritke, M. K., Rusnak, J. M., Lazo, J. S., Allan, W. P., Dive, C., Heer, S., of expression by bis-benzimidazole drugs. Mol. Pharmacol., 59: 699–706, and Yalowich, J. C. Differential induction of etoposide-mediated apopto- 2001. sis in human leukemia HL-60 and K562 cells. Mol. Pharmacol., 46: 605– 36. Wang, H., Jiang, Z., Wong, Y. W., Dalton, W. S., Futscher, B. W., and 611, 1994. Chan, V. T. Decreased CP-1 (NF-Y) activity results in transcriptional 19. Kamesaki, S., Kamesaki, H., Jorgensen, T. J., Tanizawa, A., Pommier, down-regulation of topoisomerase II␣ in a doxorubicin-resistant variant of Y., and Cossman, J. bcl-2 protein inhibits etoposide-induced apoptosis human multiple myeloma RPMI 8226. Biochem. Biophys. Res. Commun., through its effects on events subsequent to topoisomerase II-induced 237: 217–224, 1997. DNA strand breaks and their repair. Cancer Res., 53: 4251–4256, 1993. 37. Falck, J., Jensen, P. B., and Sehested, M. Evidence for repressional 20. Lang, A. J., Mirski, S. E., Cummings, H. J., Yu, Q., Gerlach, J. H., and role of an inverted CCAAT box in cell cycle-dependent transcription of the Cole, S. P. Structural organization of the human TOP2A and TOP2B human DNA topoisomerase II␣ gene. J. Biol. Chem., 274: 18753–18758, genes. Gene (Amst.), 221: 255–266, 1998. 1999. 21. Sng, J. H., Heaton, V. J., Bell, M., Maini, P., Austin, C. A., and Fisher, 38. Shibao, K., Takano, H., Nakayama, Y., Okazaki, K., Nagata, N., Izumi, L. M. Molecular cloning and characterization of the human topoisomerase H., Uchiumi, T., Kuwano, M., Kohno, K., and Itoh, H. Enhanced coexpres- II␣ and II␤ genes: evidence for isoform evolution through gene duplication. sion of YB-1 and DNA topoisomerase II␣ genes in human colorectal Biochim. Biophys. Acta, 1444: 395–406, 1999. carcinomas. Int. J. Cancer, 83: 732–737, 1999. 22. Tsai-Pflugfelder, M., Liu, L. F., Liu, A. A., Tewey, K. M., Whang-Peng, 39. Mo, Y. Y., Wang, Q., and Beck, W. T. Down-regulation of topoisomer- J., Knutsen, T., Huebner, K., Croce, C. M., and Wang, J. C. Cloning and ase II␣ in CEM cells selected for merbarone resistance is associated with sequencing of cDNA encoding human DNA topoisomerase II and local- reduced expression of Sp3. Cancer Res., 57: 5004–5008, 1997. ization of the gene to chromosome region 17q21–22. Proc. Natl. Acad. 40. Sandri, M. I., Isaacs, R. J., Ongkeko, W. M., Harris, A. L., Hickson, Sci. USA, 85: 7177–7181, 1988. I. D., Broggini, M., and Vikhanskaya, F. p53 regulates the minimal pro- 23. Jenkins, J. R., Ayton, P., Jones, T., Davies, S. L., Simmons, D. L., moter of the human topoisomerase II␣ gene. Nucleic Acids Res., 24: Harris, A. L., Sheer, D., and Hickson, I. D. Isolation of cDNA clones 4464–4470, 1996. encoding the ␤ isozyme of human DNA topoisomerase II and localisation 41. Wang, Q., Zambetti, G. P., and Suttle, D. P. Inhibition of DNA topoi- of the gene to chromosome 3p24. Nucleic Acids Res., 20: 5587–5592, somerase II␣ gene expression by the p53 tumor suppressor. Mol. Cell. 1992. Biol., 17: 389–397, 1997. 24. Qiu, J., Catapano, C. V., and Fernandes, D. J. Formation of topoi- 42. Shapiro, P. S., Whalen, A. M., Tolwinski, N. S., Wilsbacher, J., Fro- somerase II␣ complexes with nascent DNA is related to VM-26-induced elich-Ammon, S. J., Garcia, M., Osheroff, N., and Ahn, N. G. Extracellular cytotoxicity. Biochemistry, 35: 16354–16360, 1996. signal-regulated kinase activates topoisomerase II␣ through a mechanism 25. Brown, G. A., McPherson, J. P., Gu, L., Hedley, D. W., Toso, R., independent of phosphorylation. Mol. Cell. Biol., 19: 3551–3560, 1999. Deuchars, K. L., Freedman, M. H., and Goldenberg, G. J. Relationship of 43. Chen, G., Templeton, D., Suttle, D. P., and Stacey, D. W. Ras stim- DNA topoisomerase II␣ and ␤ expression to cytotoxicity of antineoplastic ulates DNA topoisomerase II␣ through MEK: a link between oncogenic agents in human acute lymphoblastic leukemia cell lines. Cancer Res., 55: signaling and a therapeutic target. Oncogene, 18: 7149–7160, 1999. 78–82, 1995. 44. Matsuo, K., Kohno, K., Sato, S., Uchiumi, T., Tanimura, H., Yamada, 26. Dereuddre, S., Delaporte, C., and Jacquemin-Sablon, A. Role of Y., and Kuwano, M. Enhanced expression of the DNA topoisomerase II topoisomerase II␤ in the resistance of 9-OH-ellipticine-resistant Chinese gene in response to heat shock stress in human epidermoid cancer KB hamster fibroblasts to topoisomerase II inhibitors. Cancer Res., 57: 4301– cells. Cancer Res., 53: 1085–1090, 1993. 4308, 1997. 45. Goswami, P. C., Roti Roti, J. L., and Hunt, C. R. The cell cycle- 27. Kimura, K., Saijo, M., Ui, M., and Enomoto, T. Growth state- and cell coupled expression of topoisomerase II␣ during S phase is regulated by cycle-dependent fluctuation in the expression of two forms of DNA to- mRNA stability and is disrupted by heat shock or ionizing radiation. Mol. poisomerase II and possible specific modifications of the higher molecular Cell. Biol., 16: 1500–1508, 1996. weight form in the M phase. J. Biol. Chem., 269: 1173–1176, 1994. 46. Boffa, L. C., Vidali, G., Mann, R. S., and Allfrey, V. G. Suppression of 28. Sullivan, D. M., Latham, M. D., and Ross, W. E. Proliferation-depen- histone deacetylation in vivo and in vitro by sodium butyrate. J. Biol. dent topoisomerase II content as a determinant of antineoplastic drug Chem., 253: 3364–3366, 1978.

47. Struhl, K. Histone acetylation and transcriptional regulatory mecha- 67. Hassan, H. T., and Rees, J. Differentiation induction therapy of my- nisms. Genes Dev., 12: 599–606, 1998. elodysplastic syndromes. Leuk. Res., 13: 633–637, 1989. 48. Grunstein, M. Histone acetylation in chromatin structure and tran- 68. Novogrodsky, A., Dvir, A., Ravid, A., Shkolnik, T., Stenzel, K. H., scription. Nature (Lond.), 389: 349–352, 1997. Rubin, A. L., and Zaizov, R. Effect of polar organic compounds on leuke- 49. McMurry, M. T., and Krangel, M. S. A role for histone acetylation in the mic cells. Butyrate-induced partial remission of acute myelogenous leu- developmental regulation of V(D)J recombination. Science (Wash. DC), kemia in a child. Cancer (Phila.), 51: 9–14, 1983. 287: 495–498, 2000. 69. Frankel, S. R., Eardley, A., Heller, G., Berman, E., Miller, W. H., Jr., Dmitrovsky, E., and Warrell, R. P., Jr. All-trans retinoic acid for acute 50. Mullen, J. R., Kayne, P. S., Moerschell, R. P., Tsunasawa, S., Gribs- promyelocytic leukemia. Results of the New York Study. Ann. Intern. kov, M., Colavito-Shepanski, M., Grunstein, M., Sherman, F., and Med., 120: 278–286, 1994. Sternglanz, R. Identification and characterization of genes and mutants for an N-terminal acetyltransferase from yeast. EMBO J., 8: 2067–2075, 1989. 70. Newmark, H. L., and Young, C. W. Butyrate and phenylacetate as differentiating agents: practical problems and opportunities. J. Cell Bio- 51. Hochhauser, D., Stanway, C. A., Harris, A. L., and Hickson, I. D. chem. Suppl., 22: 247–253, 1995. Cloning and characterization of the 5Ј-flanking region of the human topoisomerase II␣ gene. J. Biol. Chem., 267: 18961–18965, 1992. 71. Collins, A. F., Pearson, H. A., Giardina, P., McDonagh, K. T., Brusilow, S. W., and Dover, G. J. Oral sodium phenylbutyrate therapy in homozy- 52. Dwyer-Nield, L. D., Thompson, J. A., Peljak, G., Squier, M. K., Barker, gous ␤ thalassemia: a clinical trial. Blood, 85: 43–49, 1995. T. D., Parkinson, A., Cohen, J. J., Dinsdale, D., and Malkinson, A. M. Selective induction of apoptosis in mouse and human lung epithelial cell 72. Maestri, N. E., Brusilow, S. W., Clissold, D. B., and Bassett, S. S. lines by the tert-butyl hydroxylated metabolite of butylated hydroxytolu- Long-term treatment of girls with ornithine transcarbamylase deficiency. ene: a proposed role in tumor promotion. Toxicology, 130: 115–127, 1998. N. Engl. J. Med., 335: 855–859, 1996. 53. Kaufmann, S. H., Desnoyers, S., Ottaviano, Y., Davidson, N. E., and 73. Redonnet-Vernhet, I., Rouanet, F., Pedespan, J. M., Hocke, C., and Poirier, G. G. Specific proteolytic cleavage of poly(ADP-ribose) polymer- Parrot, F. A successful pregnancy in a heterozygote for OTC deficiency ase: an early marker of chemotherapy-induced apoptosis. Cancer Res., treated with sodium phenylbutyrate. Neurology, 54: 1008, 2000. 53: 3976–3985, 1993. 74. Liu, L., Shack, S., Stetler-Stevenson, W. G., Hudgins, W. R., and 54. Leder, A., and Leder, P. Butyric acid, a potent inducer of erythroid Samid, D. Differentiation of cultured human melanoma cells induced by differentiation in cultured erythroleukemic cells. Cell, 5: 319–322, 1975. the aromatic fatty acids phenylacetate and phenylbutyrate. J. Investig. Dermatol., 103: 335–340, 1994. 55. Asano, T., An, T., Mayes, J., Zwelling, L. A., and Kleinerman, E. S. Transfection of human topoisomerase II␣ into etoposide-resistant cells: 75. Perey, L., Hayes, D. F., and Kufe, D. Effects of differentiating agents transient increase in sensitivity followed by down-regulation of the en- on cell surface expression of the breast carcinoma-associated DF3-P dogenous gene. Biochem. J., 319 (Pt. 1): 307–313, 1996. epitope. Cancer Res., 52: 6365–6370, 1992. 56. Martins, L. M., Mesner, P. W., Kottke, T. J., Basi, G. S., Sinha, S., 76. Yabushita, H., and Sartorelli, A. C. Effects of sodium butyrate, dim- Tung, J. S., Svingen, P. A., Madden, B. J., Takahashi, A., McCormick, ethylsulfoxide and dibutyryl cAMP on the poorly differentiated ovarian D. J., Earnshaw, W. C., and Kaufmann, S. H. Comparison of caspase adenocarcinoma cell line AMOC-2. Oncol. Res., 5: 173–182, 1993. activation and subcellular localization in HL-60 and K562 cells undergoing 77. Khan, M. Z., Freshney, R. I., McNicol, A. M., and Murray, A. M. etoposide-induced apoptosis. Blood, 90: 4283–4296, 1997. Induction of phenotypic changes in SCLC cell lines in vitro by hexameth- 57. Chow, K. C., and Ross, W. E. Topoisomerase-specific drug sensitivity ylene bisacetamide, sodium butyrate, and cyclic AMP. Ann. Oncol., 4: in relation to cell cycle progression. Mol. Cell. Biol., 7: 3119–3123, 1987. 499–507, 1993. 78. Archer, S. Y., Meng, S., Shei, A., and Hodin, R. A. p21(WAF1) is 58. Aoyama, M., Grabowski, D. R., Isaacs, R. J., Krivacic, K. A., Rybicki, required for butyrate-mediated growth inhibition of human colon cancer L. A., Bukowski, R. M., Ganapathi, M. K., Hickson, I. D., and Ganapathi, R. cells. Proc. Natl. Acad. Sci. USA, 95: 6791–6796, 1998. Altered expression and activity of topoisomerases during all-trans retinoic acid-induced differentiation of HL-60 cells. Blood, 92: 2863–2870, 1998. 79. Saito, A., Yamashita, T., Mariko, Y., Nosaka, Y., Tsuchiya, K., Ando, T., Suzuki, T., Tsuruo, T., and Nakanishi, O. A synthetic inhibitor of histone 59. Van Lint, C., Emiliani, S., and Verdin, E. The expression of a small deacetylase, MS-27-275, with marked in vivo antitumor activity against fraction of cellular genes is changed in response to histone hyperacety- human tumors. Proc. Natl. Acad. Sci. USA, 96: 4592–4597, 1999. lation. Gene Exp., 5: 245–253, 1996. 80. Carducci, M. A., Nelson, J. B., Chan-Tack, K. M., Ayyagari, S. R., 60. Burgess, S. M., Ajimura, M., and Kleckner, N. GCN5-dependent his- Sweatt, W. H., Campbell, P. A., Nelson, W. G., and Simons, J. W. Phe- tone H3 acetylation and RPD3-dependent histone H4 deacetylation have nylbutyrate induces apoptosis in human prostate cancer and is more distinct, opposing effects on IME2 transcription, during meiosis and dur- potent than phenylacetate. Clin. Cancer Res., 2: 379–387, 1996. ing vegetative growth, in budding yeast. Proc. Natl. Acad. Sci. USA, 96: 6835–6840, 1999. 81. Sowa, Y., Orita, T., Minamikawa-Hiranabe, S., Mizuno, T., Nomura, H., and Sakai, T. Sp3, but not Sp1, mediates the transcriptional activation 61. Candau, R., Zhou, J. X., Allis, C. D., and Berger, S. L. Histone acetyl- of the p21/WAF1/Cip1 gene promoter by histone deacetylase inhibitor. transferase activity and interaction with ADA2 are critical for GCN5 func- Cancer Res., 59: 4266–4270, 1999. tion in vivo. EMBO J., 16: 555–565, 1997. 82. Nakano, K., Mizuno, T., Sowa, Y., Orita, T., Yoshino, T., Okuyama, Y., 62. Wolffe, A. P. Histone deacetylase: a regulator of transcription. Sci- Fujita, T., Ohtani-Fujita, N., Matsukawa, Y., Tokino, T., Yamagishi, H., Oka, ence (Wash. DC), 272: 371–372, 1996. T., Nomura, H., and Sakai, T. Butyrate activates the WAF1/Cip1 gene 63. Tsai, S., Valkov, N., Yang, W. M., Gump, J., Sullivan, D., and Seto, E. promoter through Sp1 sites in a p53-negative human colon cancer cell Histone deacetylase interacts directly with DNA topoisomerase II. Nat. line. J. Biol. Chem., 272: 22199–22206, 1997. Genet., 26: 349–353, 2000. 83. Russo, T., Zambrano, N., Esposito, F., Ammendola, R., Cimino, F., 64. Johnson, C. A., Padget, K., Austin, C. A., and Turner, B. M. Deacety- Fiscella, M., Jackman, J., O’Connor, P. M., Anderson, C. W., and Appella, lase activity associates with topoisomerase II and is necessary for eto- E. A p53-independent pathway for activation of WAF1/CIP1 expression poside-induced apoptosis. J. Biol. Chem., 276: 4539–4542, 2001. following oxidative stress. J. Biol. Chem., 270: 29386–29391, 1995. 65. Adachi, N., Nomoto, M., Kohno, K., and Koyama, H. Cell-cycle reg- 84. Yu, J., Zhang, L., Hwang, P. M., Rago, C., Kinzler, K. W., and Vo- ulation of the DNA topoisomerase II␣ promoter is mediated by proximal gelstein, B. Identification and classification of p53-regulated genes. Proc. CCAAT boxes: possible involvement of acetylation. Gene (Amst.), 245: Natl. Acad. Sci. USA, 96: 14517–14522, 1999. 49–57, 2000. 85. Samid, D., Wells, M., Greene, M. E., Shen, W., Palmer, C. N., and 66. Fontana, J. A., Rogers, J. S., and Durham, J. P. The role of 13-cis- Thibault, A. Peroxisome proliferator-activated receptor ␥ as a novel target retinoic acid in the remission induction of a patient with acute promyelo- in cancer therapy: binding and activation by an aromatic fatty acid with cytic leukemia. Cancer (Phila.), 57: 209–217, 1986. clinical antitumor activity. Clin. Cancer Res., 6: 933–941, 2000.

86. Kim, M. S., Kwon, H. J., Lee, Y. M., Baek, J. H., Jang, J. E., Lee, S. W., 89. Giardina, C., Boulares, H., and Inan, M. S. NSAIDs and butyrate Moon, E. J., Kim, H. S., Lee, S. K., Chung, H. Y., Kim, C. W., and Kim, sensitize a human colorectal cancer cell line to TNF-␣ and Fas ligation: the K. W. Histone deacetylases induce angiogenesis by negative regulation of role of reactive oxygen species. Biochim. Biophys. Acta, 1448: 425–438, tumor suppressor genes. Nat. Med., 7: 437–443, 2001. 1999. 87. Shack, S., Miller, A., Liu, L., Prasanna, P., Thibault, A., and Samid, D. 90. Conway, R. M., Madigan, M. C., Billson, F. A., and Penfold, P. L. Vulnerability of multidrug-resistant tumor cells to the aromatic fatty acids Vincristine- and cisplatin-induced apoptosis in human retinoblastoma. phenylacetate and phenylbutyrate. Clin. Cancer Res., 2: 865–872, 1996. Potentiation by sodium butyrate. Eur. J. Cancer, 34: 1741–1748, 1998. 88. Niitsu, N., Kasukabe, T., Yokoyama, A., Okabe-Kado, J., Yamamoto- Yamaguchi, Y., Umeda, M., and Honma, Y. Anticancer derivative of bu- 91. Lallemand, F., Courilleau, D., Buquet-Fagot, C., Atfi, A., Montagne, tyric acid (pivalyloxymethyl butyrate) specifically potentiates the cytotox- M. N., and Mester, J. Sodium butyrate induces G2 arrest in the human icity of doxorubicin and daunorubicin through the suppression of breast cancer cells MDA-MB-231 and renders them competent for DNA microsomal glycosidic activity. Mol. Pharmacol., 58: 27–36, 2000. rereplication. Exp. Cell Res., 247: 432–440, 1999.

Downloaded from mct.aacrjournals.org on October 2, 2021. © 2001 American Association for Cancer Research. The Histone Deacetylase Inhibitor Sodium Butyrate Induces DNA Topoisomerase II α Expression and Confers Hypersensitivity to Etoposide in Human Leukemic Cell Lines 1

Ebba U. Kurz, Sara E. Wilson, Kelly B. Leader, et al.

Mol Cancer Ther 2001;1:121-131.

Updated version Access the most recent version of this article at: http://mct.aacrjournals.org/content/1/2/121

Cited articles This article cites 89 articles, 51 of which you can access for free at: http://mct.aacrjournals.org/content/1/2/121.full#ref-list-1

Citing articles This article has been cited by 16 HighWire-hosted articles. Access the articles at: http://mct.aacrjournals.org/content/1/2/121.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://mct.aacrjournals.org/content/1/2/121. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.