molecular and biologic analysis of histone deacetylase ... · victoria m. richon5, j. paul...

Small Molecule Therapeutics

Molecular and Biologic Analysis of Histone DeacetylaseInhibitors with Diverse Specificities

Andrea Newbold1, Geoffrey M. Matthews1, Michael Bots1, Leonie A. Cluse1, Christopher J.P. Clarke1,Kellie-Marie Banks1, Carleen Cullinane2, Jessica E. Bolden1, Ailsa J. Christiansen1, Ross A. Dickins3,Claudia Miccolo6, Susanna Chiocca6, Astrid M. Kral5, Nicole D. Ozerova5, Thomas A. Miller5, Joey L. Methot5,Victoria M. Richon5, J. Paul Secrist5, Saverio Minucci6,7, and Ricky W. Johnstone1,4

AbstractHistone deacetylase inhibitors (HDACi) are anticancer agents that induce hyperacetylation of histones,

resulting in chromatin remodeling and transcriptional changes. In addition, nonhistone proteins, such as the

chaperone protein Hsp90, are functionally regulated through hyperacetylation mediated by HDACis.

Histone acetylation is thought to be primarily regulated by HDACs 1, 2, and 3, whereas the acetylation

of Hsp90 has been proposed to be specifically regulated through HDAC6. We compared the molecular and

biologic effects induced by an HDACi with broad HDAC specificity (vorinostat) with agents that predom-

inantly inhibited selected class I HDACs (MRLB-223 and romidepsin). MRLB-223, a potent inhibitor of

HDACs 1 and 2, killed tumor cells using the same apoptotic pathways as the HDAC 1, 2, 3, 6, and 8 inhibitor

vorinostat. However, vorinostat induced histone hyperacetylation and killed tumor cells more rapidly than

MRLB-223 and had greater therapeutic efficacy in vivo. FDCP-1 cells dependent on the Hsp90 client protein

Bcr-Abl for survival, were killed by all HDACis tested, concomitant with caspase-dependent degradation

of Bcr-Abl. These studies provide evidence that inhibition of HDAC6 and degradation of Bcr-Abl follow-

ing hyperacetylation of Hsp90 is likely not a major mechanism of action of HDACis as had been previously

posited. Mol Cancer Ther; 12(12); 2709–21. �2013 AACR.

IntroductionAltered acetylation of the N-terminal regions of his-

tone proteins through the reciprocal enzymatic activi-ties of histone acetyltransferases (HAT) and histonedeacetylases (HDAC) can facilitate the remodeling ofchromatin and mediate changes in gene expression (1–3). In addition, HATs and HDACs can regulate the

acetylation of numerous nonhistone proteins to affecttheir function, stability, or subcellular localization (4–8).Mammalian class I HDACs (1–3, 8) are predominantlyexpressed in the nucleus, and are the major mediators ofhistone deacetylation, whereas the Class IIa HDACs (4,5, 7, 9) and Class IIb HDACs (6 and 10) either shuttlebetween the nucleus and the cytoplasm or are predom-inantly expressed in the cytoplasm and deacetylatenonhistone proteins (3, 9).

A large number HDAC inhibitors (HDACi) have beendeveloped that can suppress HDAC enzymatic activityand mediate protein acetylation (1, 2). HDACis havepleiotropic anticancer activities by their inhibition oftumor cell proliferation and survival, through their effectson tumor angiogenesis, and their ability to modulateantitumor immunity (3, 10). TwoHDACis, vorinostat andromidepsin, have been approved by the U.S. Food andDrug Administration (FDA) for the treatment of hemato-logic malignancies (11).

Using genetic mouse models of cancer, we and othershave demonstrated a direct link between HDACi-medi-ated apoptosis and therapeutic efficacy (12, 13), indicatingthat direct tumor cell killing by these agents plays afundamental role in mediating antitumor responses invivo. Importantly, the combined effects of HDACi onhistone and nonhistone proteins may be important forthe therapeutic activities of these compounds (3, 14). Forexample, HDAC6-mediated deacetylation of Hsp90 can

Authors' Affiliations: 1Gene Regulation Laboratory, 2TranslationalResearch Laboratory, Cancer Therapeutics Program, The Peter MacCal-lumCancer Institute, St. Andrews Place, EastMelbourne; 3Walter and ElizaHall Institute;4 The Sir Peter MacCallum Department of Oncology, Univer-sity of Melbourne, Parkville, Victoria, Australia; 5Merck Research Labora-tories, Boston, Massachusetts; 6Department of Experimental Oncology,European Institute of Oncology; and 7University of Milan, Milan, Italy

Note: Supplementary data for this article are available at Molecular CancerTherapeutics Online (http://mct.aacrjournals.org/).

Current address for M. Bots: Academic Medical Centre, Amsterdam, TheNetherlands; current address for C.J.P. Clarke, GlaxoSmithKline, Brussels,Belgium; current address for Ailsa J. Christiansen, Institute of Pharmaceu-tical Science, Swiss Federal Institute of Technology (ETHZ), Z€urich, Swit-zerland; current address for V.M. Richon, Sanofi Pharmaceuticals, Cam-bridge, Massachusetts; and current address for J. Paul Secrist, AstraZe-neca, Waltham, Massachusetts.

Corresponding Author: Ricky Johnstone, Peter MacCallum Cancer Cen-tre, St Andrews Place, EastMelbourne, Victoria 3002, Australia. Phone: 61-3-9656-3727; Fax: 61-3-9656-1411; E-mail:[email protected]

doi: 10.1158/1535-7163.MCT-13-0626

�2013 American Association for Cancer Research.

MolecularCancer

Therapeutics

www.aacrjournals.org 2709

on March 21, 2021. © 2013 American Association for Cancer Research. mct.aacrjournals.org Downloaded from

Published OnlineFirst October 3, 2013; DOI: 10.1158/1535-7163.MCT-13-0626

http://mct.aacrjournals.org/

regulate its protein chaperone activity. Accordingly, deg-radation of client oncoproteins such as ErbB2 and Bcr-AblthroughHDACi-mediated hyperacetylation ofHsp90 hasbeen proposed to be an important functional effectormechanism of HDACis (3).

Most HDACis have broad inhibitory activity againstclass I and II HDACs (15); however, isoform-selectiveHDACis have been developed (16). It is presently notclear if selectively targeting a single HDACwill provide asignificant advantage over broad-actingHDACis in termsof increased efficacy and/or reduced toxicity. Indeed,treatment of tumor cells with the HDAC6-selective inhib-itor tubacin did not affect tumor cell growth or survival(17). Moreover, gene deletion and knockdown studiesindicate that inhibition of a singleHDACmaynot providepronounced antitumor effects. For example, siRNA-medi-ated knockdown ofHDAC1, 2, 3, or 6 failed to affect tumorcell growthor survival (18). In a studyusingRas-mediatedtumors derived from conditionalHdac1, 2, 3, and 8 knock-outmice, combined knockout ofHdac1 andHdac2, but notof any of the Hdacs1, 2, 3, or 8 alone, was sufficient toinhibit tumor growth in vitro and in vivo (19). However,although there have been some reports of single HDACinhibition affecting tumor cell growth as transient knock-down of HDAC1 or HDAC3 suppressed HeLa cell prolif-eration (20), another group showed that transient knock-down ofHDAC1, 2, or 3decreased the rate of proliferationof colon carcinoma cell lines (21). Although these studieshint that in certain cellular contexts inactivation of a singleHDAC may have an antitumor effect, most studies usinggenetic or pharmacologic approaches to inhibit HDACfunction or delete HDAC expression indicate that target-ing at least two HDACs may be necessary to mediatepotent antitumor responses.

Hsp90 is an abundant cellular chaperone whose over-expression in tumor cells correlates with poor prognosisand drug resistance (22). As part of a multiprotein com-plex that includes Hsp70, Hsp90 binds a diverse array ofclient proteins including key oncogenic and antiapoptoticproteins, preventing their ubiquitination and proteoso-mal degradation (22, 23). Hsp90 is deacetylated byHDAC6 (24), and HDACis capable of inhibiting HDAC6induce the hyperacetylation of Hsp90, resulting in pro-teosomal degradation of Hsp90 client proteins ErbB1,ErbB2/Her2, Akt, c-Raf, Bcr-Abl, and Flt-3 (25–28).Although it has been proposed that hyperacetylation ofHsp90 and subsequent degradation of oncogenic clientproteins may be amajor mediator of the antitumor effectsof HDACis (3), there is counterevidence to refute this.In addition to the lack of antitumor effects of the HDAC6inhibitor tubacin, siRNA-mediated knockdown ofHDAC6 resulted in hyperacetylation and loss of chaper-one function of Hsp90, but had little or no effect on tumorcell growth and survival (29) Moreover, a significantnumber of HDACis that are potent mediators of tumorcell death are very weak inhibitors of HDAC6 (3, 15).

In this study, we used the recently developed HDACiMRLB-223, which preferentially inhibits recombinant

HDAC1 and 2 at low nanomolar concentrations, to deter-mine if selectively targeting certain class I HDACs wouldresult in similar induction of apoptosis and therapeuticefficacy as seen using the broad-spectrum HDACi vor-inostat. We show that although MRLB-223 can induceapoptosis in B-cell lymphomas derived from Em-myctransgenic mice and can provide a therapeutic benefit tomice bearing Em-myc lymphomas, the kinetics of histonehyperacetylation and tumor cell apoptosis inducedby thiscompound is delayed, and its potency in vivo is inferior tothat observedusingvorinostat. In adifferent experimentalsetting using a panel of primary breast cancer cells, wefound that there was incomplete overlap in sensitivity tovorinostat and MRLB-223, suggesting that moleculardeterminants of sensitivity to HDACis with differentspecificity exist within distinct tumor samples. Interest-ingly, MRLB-223 was capable of killing an interleukin 3(IL-3)–dependent FDCP-1 mouse myeloid cell line thatwas engineered to grow independently of IL-3 throughforced expression of the Hsp90 client protein Bcr-Abl.Moreover, bothMRLB-223 and romidepsin, HDACis thatpotently target class I HDACs but not HDAC6, induceddegradation of Bcr-Abl and hyperacetylation of Hsp90.Furthermore, the degradation of Bcr-Abl was caspasedependent and appeared to largely occur as a down-stream consequence of tumor cell apoptosis. These dataprovide evidence that inhibition of HDAC6 and concom-itant degradation of Bcr-Abl following hyperacetylationof Hsp90 is not necessary for HDACis to kill cells that are"addicted" to the expression ofHsp90 client oncoproteins.Our data provide important insights into themechanismsof action of small molecules with diverse HDAC specifi-cities. In addition, our studies bring into question thehypothesis that degradation of Hsp90 client proteinsthrough hyperacetylation of Hsp90 following HDAC6inhibition is a major effector mechanism of action ofHDACi that target multiple HDACs.

Materials and MethodsCell culture

Em-myc, Em-myc/Bcl-2, and Em-myc/p53�/� lympho-mas were developed and grown in vitro as describedpreviously (13). The IL-3–dependent mouse bone mar-row-derived cell line FDCP-1 has been described previ-ously (30). FDCP-1 cells were cultured in RPMI1640 andconditionedmedia from theWEHI-3B cell line as a sourceof IL-3, as previously outlined (30). The FDCP-1/Bcr-Abland FDCP-1/Bcr-Abl(T315I) cells were engineered byretroviral transduction using MSCV-Bcr-Abl(p210) puroor MSCV-Bcr-Abl(T315I) puro (a gift from Scott Lowe,CSHL) and transduction methods that were previouslydescribed (13). Verification of cell lines used in thisstudy was performed using a number of in-house bio-logic assays. For example, in vitro assays on FDCP-1cells were performed to verify the requirement of IL-3for the survival of this factor-dependent cell line. West-ern blot assays were performed to verify the expression

Newbold et al.

Mol Cancer Ther; 12(12) December 2013 Molecular Cancer Therapeutics2710




or loss of appropriate genes in "compound mutant" Em-myc lymphomas. For further authentication of cells,transplantation and rejection studies of Em-myc andFDCP-1 cells were performed in genetically pure inbredstrains of recipient mice to verify the genetic origin ofthe cells.

Gene knockdown using RNA interferenceTargeting sequences for short hairpin RNA (shRNA)-

miR30 constructs against murine HDAC6 (RefSeqNM_001130416) were identified using the previouslydescribedDesigner of Small Interfering RNA (DSIR) algo-rithm (31). The top-ranked shRNAswere used to create 10miR30 sequences (97mer, Sigma-Aldrich) and these werecloned into the pLMS plasmid (MSCV)LTR-miR30-SV40-GFP vector following the generation of approximately110-bp shRNA-miR30s by amplification of 97mers using50mir30-XhoI (CAGAAGGCTCGAGAAGGTATATTGC-TGTTGACAGTGAGCG) and 50miR30-EcoRI (CTAAAG-TAGCCCCTTGAATTCCGAGGCAGTAGGCA)primers.The miRNA oligomer sequences are in the format: mir-30context-sense (underlined)-loop-antisense (underlined).HDAC 6shRNAmiR30#1. TGCTGTTGACAGTGAG-

CGCAAAGTTGGAGTTGCTACAAAATAGTGAAGCC-ACAGATGTATTTTGTAGCAACTCCAACTTTATGCC-TACTGCCTCGGAHDAC 6shRNAmiR30#2. TGCTGTTGACAGTGAG-

CGCGAGGATGACCCTAGTGTATTATAGTGAAGCC-ACAGATGTATAATACACTAGGGTCATCCTCATGC-CTACTGCCTCGGAScrambledshRNAmiR30. TGCTGTTGACAGTGAG-

CGATCTCGCTTGGGCGAGAGTAAGTAGTGAAGCC-ACAGATGTACTTACTCTCGCCCAAGCGAGAGTGC-CTACTGCCTCGGAClones were verified by sequencing using the Mouse

Stem Cell Virus (MSCV) forward primer (CCCTTGAAC-CTCCTCGTTCGACC).

ReagentsVorinostat and MRLB-223 (Merck) were dissolved in

dimethyl sulfoxide (DMSO) for the preparation of 10mmol/L stock solutions. Imatinib mesylate (Novartis)was dissolved in PBS for the preparation of a 100mmol/L stock, and romidepsin (Celgene) was suppliedas a 5-mg/mLsolution anddiluted inPBSormedia beforeuse. The structures of vorinostat, MRLB-223, romidepsin,and imatinibmesylate used in this study are shown in Fig.1A and Supplementary Fig. S1. Tetramethylrhodamineethyl (TMRE) ester was purchased from MolecularProbes, propidium iodide (PI) solution was purchasedfrom Sigma, and Annexin V was purchased from BDPharMingen. [18F]fluoro-2-deoxy-D-glucose (FDG) waspurchased from Cyclotek.

In vivo experimentsSix- to eight-week-old C57BL/6 mice were used for in

vivo apoptosis assays (IVA), small-animal positron emis-sion tomography (SA-PET), and therapy studies. The

Peter MacCallum Cancer Centre Animal Ethics Commit-tee approved all mouse protocols used in this study. Six-to eight-week-old C57BL/6 mice were injected intrave-nously with 5 � 105 Em-myc lymphomas. After the lym-phomas became palpable, mice were used for IVAs (3mice/group), SA-PET (3 mice per group), and therapystudies (10 mice/group). Therapy experiments wereundertaken using MRLB-223 (15 mg/kg daily, intraper-itoneal) and vorinostat (200mg/kg i.p for 7 days, then 150mg/kg i.p daily thereafter as previously optimized;ref. 13). Briefly, C57BL/6 mice were injected intravenous-ly with 5 � 105 Em-myc lymphomas. Peripheral whiteblood cell counts (WBC) were monitored, and therapycommenced when counts exceeded 13 � 103 cells permicroliter (Sysmex Hematology Analyzer K-1000; Sys-mex). For IVAs, C57BL/6 mice were injected with 5 �105 Em-myc lymphomas and, after 10 to 15 days duringwhich lymph nodes became well palpable, either a 200-mg/kg dose of vorinostat or a 15-mg/kg dose of MRLB-223 was administered intraperitoneally. After the indicat-ed time points, mice were sacrificed, and cells wereharvested from brachial lymph nodes for fluorescence-activated cell sorting (FACS)–based assays to measureapoptotic signaling (13). For SA-PET analyses, mice wereinjected intraperitoneallywith either a 200-mg/kg dose ofvorinostat or a 15-mg/kg dose of MRLB-223, and thenscanned on a small-animal PET scanner (Phillips) at 4, 24,and 48 hours following injection. For details, please referSupplementary Materials. For all in vivo assays, the vehi-cle used to formulate vorinostat and MRLB-223 wasDMSO. MRLB-223 was further diluted and administeredwith 50:50 (PEG400:H2O). Therefore, although the vehiclefor vorinostat was DMSO, the vehicle for MRLB-223 was50:50 (PEG400:H2O).

Pharmacokinetic analysis of vorinostat andMRLB-223

C57BL/6mice were treated by intraperitoneal injectionof either vorinostat (200mg/kg) orMRLB-223 (15mg/kg).Peripheral blood (�200 mL) was obtained from individualmice (three per time point) at 2, 4, 8, 12, and 24 hours aftertreatment. Serum was then collected, and pharmacoki-netic analysis was undertaken at the Merck ResearchLaboratories (Boston,MA). Briefly, plasma concentrationswere determined using high-performance liquid chroma-tography/tandem mass spectrometry (HPLC/MS–MS).Then, 50-ml plasma samples were precipitated using 200mL of acetonitrile containing an internal standard. Theprecipitated proteins were separated by filtration and theresulting supernatant was evaporated and then reconsti-tuted in 100 mL acetonitrile:water (30:70 v/v). A 10-mLsample was injected on a C18 column (2.0 mm� 30 mm, 3mm particle size) and eluted using a gradient LC methodwith water containing 0.1% formic acid as the aqueousmobile phase, and acetonitrile containing 0.1% formicacid as the organic phase. Electrospray ionization withmultiple reaction monitoring was used for MS–MS detec-tion. A standard curve ranging from 0.003 to 10 mmol/L

Antitumor Activities of Isoform-Selective HDAC Inhibitors

www.aacrjournals.org Mol Cancer Ther; 12(12) December 2013 2711




was prepared and used to calculate the concentration ofunknown samples.

Western blotting and immunoprecipitationWestern blot assays were performed as previously

described (13) using the following antibodies: anti-acety-lated tubulin (Sigma-Aldrich); anti-acetylated H3 and H4(Millipore); anti-b-actin (Clone AC-74, Sigma-Aldrich);and anti-c-Abl (Calbiochem). For immunoprecipitationstudies, the FDCP-1/Bcr-Abl cell lines were lysed in ice-cold immunoprecipitation lysis buffer (1% Triton X-100,150mmol/LNaCl, 2.0mmol/L EDTA, 10 mg/mL leupep-tin, 1 mg/mL Pepstatin-A, 2 mg/mL aprotinin, 25mmol/LNaF, and 0.5 mmol/L sodium orthovanadate) for 30minutes, and the nuclear and cellular debris was clearedby centrifugation. Immunoprecipitation was performed

with sepharaose protein G beads (Zymed Laboratories)incubatedwith1mgof totalproteinpluseitheranti-HSP90(Assay designs Stressgen) or normalmouse immunoglob-ulin G1 (IgG1) isotype control antibody (Santa Cruz Bio-technology) at 4�C for 16 hours. The immunoprecipitateswere washed 3 times in lysis buffer, and proteins wereeluted with the SDS sample-loading buffer. Proteins wereseparated by SDS-PAGE, and acetylated lysine proteinswere detected using an anti-acetylated lysine antibody(Cell Signaling Technology). Secondary antibody incuba-tions and chemiluminescence visualizations were per-formed as previously described (13).

In vitro cell death analysisCells (5–8 � 105 cells/mL) were incubated in the

presence of the indicated compounds for 24 to 72 hours

0

20

40

60

80

100

0 0.05 0.1 0.5 1 1.5

Per

cent

MO

MP

Vorinostat (µmol/L)

48 h

24 h

0

20

40

60

80

100

0 0.2 0.5 1 5 10

Per

cent

MO

MP

MRLB 223 (µmol/L)

48 h

24 h

A

0

20

40

60

80

100

0 0.5 1 1.5 2

Per

cent

PI p

osi

tive

Vorinostat (µmol/L)

48 h24 h

B C

E

48 hr

24 hr

48 h

24 h

0 2 4 6 8 10 12

MRLB-223 (µmol/L)

100

80

60

40

20

0

D

MRLB-223

Per

cent

PI p

osi

tive

Figure 1. In vitro sensitivity of Em-myc lymphomas to isoform-selective HDACi. A, chemical structure of MRLB-223. B–E, Em-myc lymphomas were incubatedin vitro for 24 and 48 hours with the indicated concentrations of either vorinostat or MRLB-223. Cell viability was assessed by PI staining (A and C) andTMRE (B and D) staining to assess mitochondrial outer membrane permeabilization (MOMP). Data are the mean � SEM of at least three independentexperiments.

Newbold et al.





within 1 mL cell culture media in 24-well plates (GreinerBio-One). Viability of cells as measured by PI uptake,Annexin V staining, cell-cycle analysis, or TMRE stain-ing were performed as detailed previously (32, 33).FDCP1 cells (5–8 � 105 cells/mL) were incubated withor without 50 mmol/L Q-VD-OPh in the presence of theindicated HDACi for 24 to 48 hours in 5-mL cultureflasks. Cell viability was measured by PI uptake, andcell suspensions were harvested for Western blotanalysis.

ResultsMolecular and biologic activities of broadly actingand HDAC1/2-specific HDACisMRLB-223 (34) is anHDACi that selectively inhibits the

enzymatic activity of HDAC1 and -2 (SupplementaryTable S1). In comparison, vorinostat has broader HDACinhibitory activities and is capable of inhibiting all fourclass I HDACs (1, 2, 3, and 8) as well as HDAC6 atconcentrations lower than 500 nmol/L (SupplementaryTable S1). We therefore, aimed to compare the ability ofMRLB-223 and vorinostat to kill Em-myc cells. Vorinostat

effectively killed Em-myc lymphomas in a concentration-dependent manner after a 24-hour incubation, as shownby the increase in the percentage of PI uptake (Fig. 1B) andan increase in the percentage mitochondrial outer mem-brane permeabilization (mitochondrial outer membranepermeabilization (MOMP); Fig. 1C). Following an addi-tional 24-hour exposure with vorinostat, there was littlechange in the IC70 observed. In addition, MRLB-223induced an increase in PI uptake (Fig. 1D) and anincrease in MOMP (Fig. 1E) in the Em-myc cells aftera 24-hour exposure. However, after a 48-hour exposureto the compound, the level of PI uptake mitochondrialmembrane permeabilization significantly increased,reducing the IC70 concentration from 5 to 0.5 mmol/L.As a result, using MRLB-223 over a 48-hour periodresulted in induction of cell death equivalent to thatobserved using an equivalent concentration of vorino-stat for 24 hours. In experiments using both compounds,the change in MOMP correlated very well with theuptake of PI.

Our previous studies demonstrated that vorinostatinduced tumor cell apoptosis via the intrinsic apoptosispathway (13, 35). Vorinostat-mediated cell killingwas not

0

20

40

60

80

100

0 0.5 1 1.5 2

Per

cent

PIp

osi

tive

Vorinostat (mmol/L)

0

20

40

60

80

100

6420 8 10 12

Per

cen

t P

I po

siti

ve

MRLB-223 (mmol/L)

Eµµ-myc

Eµ-myc/p53-/-

Eµ-myc/Bcl-2

Eµ-myc

Eµ-myc/p53-/-

Eµ-myc/Bcl-2

A

B

C

Ac-H3 and -H4

Ac-tubulin

b-Actin

Eµ-myc/Bcl2

Ac-H3 and -H4 Eµ-myc

Ac-tubulin

b-Actin

2 8 24 2 8 24 2 8 24

MRLB-223 Vorinostat Vehicle

Ac-H3 and -H4

Ac-tubulin

b-Actin

Eµ-myc/p53-/-

Figure 2. Apoptosis induced by MRLB-223 and vorinostat is blocked by Bcl2 but is independent of a functional p53 pathway. Em-myc, Em-myc/p53�/�,and Em-myc/Bcl-2 lymphomas were incubated in vitro for 24 hours with the indicated concentrations of (A) vorinostat or (B) MRLB-223. Cell viabilitywas assessed byPI staining. Data are themean�SEMof at least three independent experiments. C, Em-myc, Em-myc/p53�/�, and Em-myc/Bcl-2 lymphomaswere incubated in vitro with 0.5 mmol/L vorinostat or 3 mmol/L MRLB-223 for 2, 8, and 24 hours. Cells were harvested, and Western blot analysis wasperformed with antibodies to detect acetylated tubulin and acetylated histones H3 and H4. Anti-b-actin was used as a loading control.






dependent on a functional p53 pathway, and was effec-tively blocked by overexpression of prosurvival proteinsof the Bcl-2 family (13, 36). We, therefore, determined ifMRLB-223 induced tumor cell apoptosis via a similarmechanism. Em-myc/p53�/� tumor cells were sensitiveto vorinostat andMRLB-223 at similar concentrations andwith the same kinetics as observed in Em-myc lymphomas(Fig. 2A and B). In contrast, overexpression of Bcl-2suppressed tumor cell death mediated by both agents.These data indicate that, although vorinostat and MRLB-223 have a different repertoire ofHDAC targets, they bothinduce apoptosis of Em-myc cells via the intrinsic apopto-tic pathway and both function independently of p53.

To demonstrate the differences in HDAC specificitybetween vorinostat and MRLB-223 in a cellular settingand to formally show that loss of expression of p53 oroverexpression of Bcl-2 did not affect the HDAC inhibi-tory functions of the compounds, Western blot assayswere performed to assess the acetylation status of histonesand a-tubulin. Deacetylation of a-tubulin is mediated byHDAC6 and, therefore, the acetylation state of a-tubulinserves as a surrogate readout for HDAC6 activity in cell-based systems (37, 38). Both vorinostat and MRLB-223induced hyperacetylation of histonesH3 andH4 althoughthe kinetics of response were different (Fig. 2C). Hyper-acetylation of histone H3 and H4 occurred after a 2-hourexposure to vorinostat in Em-myc and Em-myc/Bcl-2 lym-phomas and reached a maximum level following 8 hoursof exposure whereas, in Em-myc/p53�/� lymphomas, themaximum level of histoneH3andH4acetylationwas seenat the 24-hour time point. In contrast, MRLB-223 inducedonly minimal histone hyperacetylation after 2-hour expo-sure, and this gradually increased over time. Vorinostatinduced hyperacetylation of a-tubulin with maximal acet-ylation thatwasobserved following 8-hour exposure to thecompound. Consistent with the HDAC1/2 binding spec-ificity of MRLB-223 shown in Supplementary Table S1,there was no increase in acetylation ofa-tubulin followingexposure of Em-myc, Em-myc/Bcl-2, or Em-myc/p53�/�

lymphomas to the compound. Taken together, these datademonstrate that both vorinostat and MRLB-223 inhibitHDACs that deacetylate histones although the kinetics ofthe MRLB-223 response were slower than that observedwith vorinostat. In addition, vorinostat, but notMRLB-223,inhibited HDAC6 as measured by the hyperacetylation ofa-tubulin.

In vivo effects of vorinostat and MRLB-223We, next, wanted to compare the antitumor effects

of vorinostat and MRLB-223 in vivo. Pharmacokineticstudies using vorinostat and MRLB-223 at their maxi-mum-tolerated doses (MTD) revealed that the maxi-mum serum concentration of MRLB-223 was signifi-cantly greater than vorinostat (Supplementary TableS2). To determine the therapeutic effect of vorinostatand MRLB-223, Em-myc lymphomas were transplantedinto C57BL/6 mice and treatment with the HDACior vehicle commenced when WBC counts reached

13 � 106/mL or greater (day 7 post lymphoma injec-tion). The WBC counts in tumor-bearing mice treatedwith vehicle increased over time to an average of39.2 � 103/mL on day 21 that increased to 54.3 by day28 (Fig. 3A). Treatment of mice with MRLB-223decreased the blood tumor load and, by day 21, theaverage WBC count was 23.9 � 103/mL and increased to33.8 � 103/mL by day 35. In contrast, treatment of micebearing Em-myc lymphomas with vorinostat reducedthe WBC counts to normal levels, and the reduction inblood tumor load was maintained throughout the vor-inostat treatment period. After treatment with vorino-stat ceased, the WBC counts increased over a 2-weekperiod to a maximum level of 31.5 � 103/mL by day 55.

Vorinostat- andMRLB-223–treated mice demonstrateda significant increase in survival compared with vehicle-treated mice (Fig. 3B; median survival vehicle ¼ 33 days,median survival MRLB-223 ¼ 42 days, median survivalvorinostat ¼ 63 days). Clearly, MRLB-223 provided atherapeutic benefit to the mice; however, the effect wasnot nearly as robust as that provided by vorinostat. Con-sistent with our in vitro data (Fig. 2C), the kinetics ofhistone hyperacetylation mediated by MRLB-223 in vivowas delayed compared with the response to vorinostat(Fig. 3C).

In addition, the more rapid and robust response tovorinostat in vivo was demonstrated by molecular imag-ing using FDG-PET. The uptake of FDG at baseline andfollowing 4- and 48-hour exposure to vorinostat andMRLB-223 is shown in Fig. 3D. At baseline, there wasspecific uptake of FDG in the lymph nodes and spleen oftumor-bearing mice as well as nonspecific uptake in theheart andbladder. TheFDGsignal in the lymphoid organsdiminished significantly following 4-hour treatment withvorinostat (Supplementary Table S3) and remained low at48 hours. In contrast, treatment with MRLB-223 had littleor no effect on FDG uptake at the 4-hour time point(Supplementary Table S3); however, uptake in the lymphnodes at 48 hours was decreased.

HDAC6 inhibition is not required to kill cellsdependent on the Hsp90 client protein Bcr-Abl

To test the hypothesis that inhibition ofHDAC6 activityand concomitant degradation of Hsp90 client oncopro-teins followinghyperacetylation ofHsp90plays an impor-tant mechanistic role in HDACi-induced cell death (25,26, 29), we expressed wild-type Bcr-Abl and a mutantform of the fusion protein (T315I) that does not respond toimatinib (39) in the IL-3–dependent cell line FDCP-1.FDCP-1/Bcr-Abl and FDCP-1/Bcr-Abl(T315I) cells grewin the absence of IL-3 whereas parental FDCP-1 cells died(Supplementary Fig. S2A). We confirmed the expressionof Bcr-Abl and Bcr-Abl(T315I) in FDCP-1 cells byWesternblot analysis (Supplementary Fig. S1B). In addition, weshowed the functional activity of Bcr-Abl by treatingFDCP-1, FDCP-1/Bcr-Abl, and FDCP-1/Bcr-Abl(T315I)cells with imatinib (Supplementary Fig. S2C). Only cellsexpressing wild-type Bcr-Abl were sensitive to imatinib

Newbold et al.





whereas the T315I substitution rendered Bcr-Abl insen-sitive to imatinib, butmaintained its prosurvival function.As with the Em-myc cells, the kinetics of MRLB-223-

induced death of FDCP-1 cells were slower than thatobserved using vorinostat (Fig. 4A and B). Vorinostatinduced hyperacetylation of histones H3 and H4 anda-tubulin following a 2-hour incubation, whereasMRLB-223 induced only acetylation of histone H3 andH4 with delayed kinetics compared with vorinostat andno acetylation of a-tubulin was observed (Fig. 4C). Inaddition, we tested the sensitivity of FDCP-1 cells toromidepsin, an HDACi that potently inhibits HDACs 1,2, and 3 (see Supplementary Table S1; ref. 15). Romidepsineffectively killed FDCP-1 cells (Fig. 4D) and inducedacetylation of histone H3 following 24-hour incubation,but did not affect acetylation of a-tubulin (Fig. 4E).

We next determined the effect of vorinostat,MRLB-223,and romidepsin on FDCP-1/Bcr-Abl and FDCP-1/Bcr-Abl(T315I) cells grown in the absence of IL-3 (Fig. 5A–Cand Supplementary Fig. S3). FDCP-1 cells expressingeither wild-type or mutant Bcr-Abl were equivalentlysensitive to the threeHDACis. Tumor cell deathmediatedby MRLB-223 required longer exposure times comparedwith vorinostat and romidepsin. The effect of differentHDACi on Bcr-Abl expression was assessed by Westernblot analysis. Treatment with 5 mmol/L vorinostat for 24hours induced degradation of Bcr-Abl concomitant withsubstantial apoptosis (Fig. 5D). In addition, treatmentwith MRLB-223 and romidepsin induced cell death and,interestingly, Bcr-Abl expression was lost concomitantwith induction of apoptosis (Fig. 5D). Moreover, inhibi-tion of HDACi-induced apoptosis using the caspase

B

Ac-H4

Ac-H3

b-Actin

Vorinostat

2 8 24 48

MRLB-223

2 8 24 48 Time (h)

C

0 20 40 60 80 100Time (d)

Time (d)

WB

C (

¥106

/mL

)

100

80

60

40

20

0

Su

rviv

al p

rob

abili

ty (

%)

VehicleMRLB-223Vorinostat

Vehicle

MRLB 223

Vorinostat

Therapy

Baseline Vorinostat (4 h) Vorinostat (48 h)

Baseline MRLB-223 (4 h) MRLB-223 (48 h)

A

C D

70

60

50

40

30

20

10

05 10 15 20 25 30 35 40 45 50 55 60

Figure 3. In vivo sensitivity of Em-myc lymphomas to isoform-selective HDACi. Em-myc lymphomas were transplanted into C57BL/6 mice, and treatment withvehicle, vorinostat, or MRLB-223 commenced when WBC counts reached 13 � 106/mL or higher. A, WBC counts from mice transplanted with Em-myclymphomas from the commencement of therapy and over timewith vehicle, vorinostat, orMRLB-223. Data are themean�SEMof at least three independentexperiments. B, Kaplan–Meier survival curves of vehicle-treatedmice (green line),MRLB-223–treatedmice (blue line), and vorinostat-treatedmice (brown line)are shown. Median survival was: 33 days for vehicle; 42 days for MRLB-223, P ¼ 0.0015; and 63 days for vorinostat, P ¼ 0.0026. There was astatistically significant difference in median survival in vorinostat- versus MRLB-223–treated mice (P ¼ 0.0050). Statistical analyses were performed usingMedCalc statistical software. C, C57Bl/6 mice bearing Em-myc lymphomas were injected with one dose of either vorinostat (200 mg/kg) or MRLB-223(15mg/kg). Lymphoma cells were harvested at various time points (hours) indicated followingHDACi treatment andWestern blot analysis was performedwithantibodies to detect acetylated H3 and acetylated H4. Anti-b-actin was used as a loading control. D, FDG-based PET uptake was measured in C57Bl/6mice bearing Em-myc lymphomas at baseline as well as 4 and 48 hours after a 200-mg/kg dose of vorinostat (top) or a 15-mg/kg dose of MRLB-223 (bottom).All time points in each panel represent the same mouse, and each panel is a single representative from three mice/group.






inhibitor Q-VD-OPh resulted in maintenance of Bcr-Ablexpression following HDACi treatment (Fig. 5E).

HDAC6 inhibition is not required for HDACi toinduce the hyperacetylation of Hsp90 in FDCP-1/Bcr-Abl cells

Our data indicated that inhibition of HDAC6 andsubsequent degradation of Hsp90 client oncoproteinssuch as Bcr-Abl was not a major apoptosis effectormechanism of HDACi action in our experimental sys-tem. The current literature suggests that HDAC6 is adirect deacetylase for HSP90 and that HDAC6 inhibi-tion leads to the hyperacetylation of Hsp90 (24, 40, 41).To investigate this in our system, we performed immu-noprecipitation/Western blotting assays to assess the

acetylation state of Hsp90 in FDCP-1/Bcr-Abl cells aftertreatment with HDACis of different isoform specificities(Fig. 6A). Vorinostat and romidepsin induced hypera-cetylation of Hsp90 following an 8-hour incubationcompared with the DMSO-treated controls. Important-ly, although MRLB-223 required a longer incubationtime of 24 hours, it was also able to induce hyperace-tylation of Hsp90. Our results indicate that direct inhi-bition of HDAC6 by HDACis is not necessary for theseagents to induce Hsp90 hyperacetylation and subse-quent degradation of client proteins.

We complemented the studies detailed in Fig. 6A usinga gene knockdown approach. FDCP-1 cells were tra-nsduced with retroviruses expressing shRNAs embe-dded within the miR30 miRNA that targeted HDAC6

0

20

40

60

80

100

1612840

24 hr48 hr

0

20

40

60

80

100

302520151050

Perc

ent P

Ipos

itive

MRLB-223 (mmol/L)

A

0

20

40

60

80

100

86420Vorinostat (mmol/L)

Perc

ent P

Ipos

itive 24 h

48 h

24 h48 h72 h

MRLB-223 elciheVtatsoniroV2 8 24 2 8 24 2 8 24

Ac-H3 and -H4

Ac-tubulin

b-Actin

B

C D

24 h48 h

Perc

ent P

I upt

ake

Romidepsin (nmol/L)

Ac-tubulin

Ac-H3

b-Actin

Time (h): 0 24 0 24

Vorinostat RomidepsinE

Figure 4. In vitro sensitivity of FDCP1 cells to HDACi. FDCP-1 cells were incubated with the indicated concentrations of: A, vorinostat; B, MRLB-223;or D, romidepsin; and time points were taken at 24, 48, and/or 72 hours. Cell viability was assessed by PI staining, and data are the mean � SEMof at least three independent experiments. In addition, FDCP1 cells were incubated in vitro with (C) 3 mmol/L vorinostat and 5 mmol/L MRLB-223 for2, 8, and24hours; (C) or (E) 3mmol/L vorinostat and3nmol/L romidepsin for 0 and24hours. The cellswere harvested andWestern blot analysiswasperformedwith antibodies that detect acetylated tubulin and acetylated histones H3 and H4. Anti-b-actin was used as a loading control.

Newbold et al.





(HDAC6shRNAmiR#1 and HDAC6shRNAmiR#2) orwith a control scrambled shRNA (scrambledshRNAmiR).FDCP-1/Bcr-Abl/HDAC6shRNAmiR#1, FDCP-1/Bcr-Abl/HDAC6shRNAmiR#2 and FDCP-1/Bcr-Abl/scram-bledshRNAmiR cells were viable and analyzed for exp-ression of HDAC6 and tubulin acetylation. Both FDCP-1cell lines expressing HDAC6shRNAmiRs had knock-down of HDAC6 to below detectable levels and concom-itant hyperacetylation of a-tubulin (Fig. 6B). Consistentwith previous data using a specific HDAC6 inhibitor (17),knockdown of HDAC6 did not induce histone acetylation(Supplementary Fig. S4A). However, knockdown ofHDAC6 resulted in hyperacetylation of Hsp90 (Fig.6C). Interestingly, and in contrast to the expected result,

the expression of Bcr-Abl was increased in FDCP-1/Bcr-Abl/HDAC6shRNAmiR#1 and FDCP-1/Bcr-Abl/HDAC6shRNAmiR#2 (Fig. 6D and Supplementary Fig.S4B). Our data, therefore, indicate that inhibition ofHDAC6 activity or loss of HDAC6 expression is suffi-cient to induce hyperacetylation of Hsp90 but not nec-essary nor sufficient to cause a decrease in expression ofBcr-Abl.

To formally determine if inhibition of HDAC6 wasimportant for mediating the apoptotic effects of a multi-HDAC inhibitor such as vorinostat or if loss of HDAC6could functionally interact with a more selective HDACisuch as MRLB-223, we tested FDCP-1/Bcr-Abl/HDA-C6shRNAmiR#1,FDCP-1/Bcr-Abl/HDAC6shRNAmiR#2,

FDCP1/Bcr-Abl(T315I)FDCP1/Bcr-Abl

FDCP1

0 2 4 6 8 10 12 14

100

80

60

40

20

0

Per

cent

PI p

ositi

ve

Romidepsin (nmol/L)

0

20

40

60

80

100

Per

cen

t P

I po

siti

ve

0

20

40

60

80

100


Per

cent

PI p

ositi

ve

A B

FDCP1/Bcr-Abl(T315I)FDCP1/Bcr-AblFDCP1

FDCP1/Bcr-Abl(T315I)FDCP1/Bcr-AblFDCP1

C

9% : 61%B

cr-A

bl

+5 m

mo

l/L V

ori

no

stat

Bcr-Abl

b-Actin

10% : 33% : 42% : 20%

Bcr-Abl

b-Actin

Bcr

-Ab

l

+5 m

mo

l/L 2

23

+10

mmo

l/L 2

23

+3 n

mo

l/L r

om

idep

sin

EBcr-Abl

b-Actin

Percent PI positive: 9.49, 38.3, 7.93, 7.25

Q-VD-OPh: - - + +5 mmol/L vorinostat: - + - +

0 5 10 15 20 25

MRLB-223 (mmol/L)

Percent PI positive: 10.4, 32.9, 42.0, 19.9, 4.5 6.2, 10.4, 11.6

Bcr-Abl

b-Actin

Q-VD-OPh: - - - - + + + +5 mmol/L MRLB-223: - + - - - + - -

10 mmol/L MRLB-223: - - + - - - + -3 nmol/L Romidepsin: - - - + - - - +

D

Figure 5. HDACis with different isoform specificities induce degradation of Bcr-Abl–expressing cells after death. FDCP-1, FDCP-1/Bcr-Abl, and FDCP-1/Bcr-Abl(T315I) cells were incubated with the indicated concentrations of (A) vorinostat (24 hours), (B) MRLB-223 (72 hours), or (C) romidepsin (48 hours).Cell viability was assessed by PI staining, and data are the mean� SEM of at least three independent experiments. D, FDCP-1/Bcr-abl cells were incubatedwith 5 mmol/L vorinostat for 24 hours (left), or 5 and 10 mmol/L of MRLB-223 and 3 nmol/L romidepsin for 48 hours (right). The cells were harvested,cell viability was assessed by PI staining (shown as percentage of PI-positive cells at the bottom of the Western blots), and degradation of Bcr-Abl wasanalyzed via Western blot. Expression of b-actin was used as a loading control. The results are representative of at least three independent experiments.E, FDCP-1/Bcr-Abl cells were incubated for 24 hourswith 5mmol/L vorinostat with or without 50 mmol/L Q-VD-OPh (left), or 48 hourswith either 5 or 10 mmol/LMRLB-223, and 3 nmol/L romidepsin, with or without 50 mmol/L Q-VD-OPh (right). The cells were harvested, cell viability was assessed by PI staining,and the degradation of Bcr-Abl was analyzed via Western blot. Expression of b-actin was used as a loading control.






and FDCP-1/Bcr-Abl/scrambledshRNAmiR cells forsensitivity to vorinostat and MRLB-223. As shown in Fig.6E and F, decreased expression of HDAC6 to belowdetectable levels neither suppressed the apoptotic effectsof vorinostat nor enhanced the apoptotic effects of MRLB-223.

DiscussionHDACis are a new class of anticancer agents that are

gaining widespread use in the clinic for the treatment ofhematologic malignancies (42). There are two HDACi forcutaneous T-cell lymphoma (CTCL) that have beenapproved by the FDA—the Merck compound vorinostat

(SAHA, Zolinza) and the Celgene agent romidepsin (dep-sipeptide, Istodax). At least 16 other companies haveHDACis in clinical trials, mostly as oncology agents, andthe majority of these compounds are somewhat nonse-lective, inhibiting more than 1 of the 11 class I, II, and IVHDACs (3). For example, vorinostat equivalently inhibitsHDACs 1, 2, 3, and 6, whereas romidepsin has very highaffinity for HDACs 1, 2, and 3 and more weakly inhibitsHDAC4 (15). In addition to histones, the function, local-ization, or stability of a number of nonhistone proteins canbe regulated by acetylation (5) and, thus, the combinedeffects of HDACi on histone and nonhistone proteinsmaybe important for the therapeutic activities of these com-pounds (14). For example, the protein chaperone activity

Vehi

cle

Ac-lysine

hsp90

IP: hsp90IB:Vo

rinos

tat

Vehi

cle

MR

LB-2

23

Vehi

cle

Rom

idep

sin

A

HDAC6

b-Actin

1 2 3 4

Bcr-Abl

b-Actin

1 2 3 4

Ac-tubulin

b-Actin

1 2 3 4

Acetyl lysine

HSP90

1 2IP: HSP90IB:

B

C

D

0

20

40

60

80

100


Perc

ent P

I pos

itive untransduced

scrambled6.2KD

E

FDCP1/Bcr-AblFDCP1/Bcr-Abl/scrambledshRNAmiRFDCP1/Bcr-Abl/HDAC6shRNAmiR#1

0

20

40

60

80

100

1614121086420MRLB-223 (mmol/L)

Perc

ent P

I pos

itive

untransducedscrambled6.2KD

FDCP1/Bcr-AblFDCP1/Bcr-Abl/scrambledshRNAmiRFDCP1/Bcr-Abl/HDAC6shRNAmiR#1

F

Figure 6. Regulated acetylation of Hsp90 and expression of Bcr-Abl by HDACis and knockdown of HDAC6. A, whole-cell lysates from FDCP-1/Bcr-Ablcells treated with 10 mmol/L vorinostat for 8 hours, 10 mmol/L MRLB-223 for 24 hours, or 3 nmol/L romidepsin for 8 hours were immunoprecipitatedwith an anti-Hsp90 antibody and Western blot assays for acetylated proteins was performed. The blot was stripped and re-probed for expression of Hsp90.Results are representative of at least three independent experiments. B,Western blot analyseswere performed using lysates fromanFDCP-1/Bcr-Abl (lane 1),FDCP-1/Bcr-Abl/scrambledshRNA-miR30 (lane 2), FDCP-1/Bcr-Abl/HDAC6shRNA-miR30#1 (lane 3), and FDCP1/Bcr-Abl/HDAC6shRNA-miR30#2 (lane 4)with antibodies that detect HDAC6 (left) and acetylated tubulin (right). The expression of b-actin was used as a loading control. C, whole-celllysates from FDCP-1/Bcr-Abl/scrambledshRNA-miR30 (lane 1) and FDCP-1/Bcr-Abl/HDAC6shRNA-miR30#1 (lane 2) cells were immunoprecipitatedwith an anti-HSP90 antibody and Western blot assays for acetylated proteins was performed. The blot was stripped and re-probed for expressionof Hsp90. Results are representative of at least three independent experiments. D, Western blot analyses were performed using lysates from anFDCP-1/Bcr-Abl (lane 1), FDCP-1/Bcr-Abl/scrambledshRNA-miR30 (lane 2), FDCP-1/Bcr-Abl/HDAC6shRNA-miR30#1 (lane 3), and FDCP-1/Bcr-Abl/HDAC6shRNA-miR30#2 (lane 4) with antibodies that detect Bcr-Abl. The expression of b-actin was used as a loading control. E and F, FDCP-1/Bcr-Abl,FDCP-1/Bcr-Abl/scrambledshRNA-miR30, and FDCP-1/Bcr-Abl/HDAC6shRNA-miR30#1 cells were incubated with the indicated concentrations ofvorinostat or MRLB-223 for 48 and 72 hours, respectively. Cell viability was assessed by PI staining. Data are the mean� SEM of at least three independentexperiments.

Newbold et al.





of Hsp90 can be regulated through HDAC6-mediateddeacetylation of Hsp90. Accordingly, degradation of cli-ent oncoproteins such as ErbB2 and Bcr-Abl throughHDACi-mediated hyperacetylation of Hsp90 has alsobeen proposed to be a major functional effector mecha-nism of HDACi action (3).It is not yet known if single HDAC isoform-selective

inhibitors would provide better exposure and higherdose levels in vivo, which might lead to better efficacy inthe clinic. In order to answer this question and determineif inhibition of HDAC6 and subsequent hyperacetylationof Hsp90 and degradation of client proteins such as Bcr-Abl was a major mechanism of action of HDACi, wecompared and contrasted the molecular, biologic, andtherapeutic activities of two compounds, vorinostatand MRLB-223, that possess different HDAC inhibitoryprofiles. Specifically, we aimed to determine if inhibitionof HDACs 1 and 2 by MRLB-223 had the same apoptoticand therapeutic effects as vorinostat, which can inhibitHDACs1, 2, 3, and 6 (SupplementaryTable S1; ref. 15).Weshowed that both agents induced tumor cell death in ap53-independent manner that was inhibited by overex-pression of Bcl-2. This is consistent with our previousstudies and those of others showing that HDACis induceapoptosis via the intrinsic apoptotic pathway (13, 15,32, 35, 43–50).Consistent with the HDAC-specific inhibitory prop-

erties of the two agents, both vorinostat and MRLB-223induced histone hyperacetylation; however, only vor-inostat induced acetylation of a-tubulin, a functionknown to be regulated by HDAC6 (37). Interestingly,the kinetics of histone hyperacetylation and subsequenttumor cell apoptosis induced by MRLB-223 wasmuch slower than vorinostat. Histone acetylation canbe regulated by multiple class I HDACs, and the abilityof vorinostat to inhibit HDAC3 may confer a morerobust and rapid histone hyperacetylation effect thanthat observed with MRLB-223. Consistent with thishypothesis, HDAC3 is a known regulator of histoneacetylation (51). Moreover, romidepsin specifically inhi-bits HDACs 1, 2, and 3 (15), and we have previouslydemonstrated that vorinostat and romidepsin inducehistone hyperacetylation and tumor cell death with verysimilar kinetics (32). Taken together, these data dem-onstrated that although the multi-HDACi vorinostatinduced histone acetylation and killed tumor cells withfaster kinetics than the HDAC 1/2-selective inhibitorMRLB-223, both agents appeared to mediate their bio-logic effects through the same molecular pathway—inthis case, the intrinsic apoptotic pathway. Whetherthe delayed kinetics of MRLB-223 is the result of itsunique HDACi specificity or due to another character-istic, such as slower cell penetration, remains to bedetermined.The differences in antitumor potency and kinetics of

histone hyperacetylation observed between vorinostatand MRLB-223 in vitro were also apparent when theagents were tested in vivo in mice bearing Em-myc lym-

phomas. AlthoughMRLB-223 reached higher serum con-centrations and was more persistent at the MTD thanvorinostat, the therapeutic effects of the agent were infe-rior to those observed using vorinostat. These data implythat selective inhibition of HDACs 1 and 2 may not be theoptimal strategy for the clinical development of morepotent and tolerable HDACi.

To determine the importance of HDAC6 inhibition inmediating the antitumor effects of HDACi, we estab-lished an "oncogene-addiction" model by expressingBcr-Abl in IL-3–dependent FDCP-1 cells and growingthe cells in the absence of IL-3. We showed that vor-inostat, MRLB-223, and romidepsin were all capable ofkilling FDCP-1 cells that overexpressed wild-type or animatinib-insensitive form of Bcr-Abl (Bcr-Abl/T315I),and that MRLB-223 showed a similar delay in inducinghistone hyperacetylation and tumor cell apoptosis asobserved using Em-myc lymphomas. Interestingly, weshowed that all three HDACi could induce degradationof Bcr-Abl, but this effect was suppressed by cotreat-ment with a caspase inhibitor. These data indicate thatdegradation of the Hsp90 client protein Bcr-Abl occursas a result of induction of apoptosis pathways and isnot a primary mediator of HDACi-induced tumor celldeath as has been previously proposed. Moreover, weshowed that Hsp90 was hyperacetylated followingtreatment with romidepsin and MRLB-223 althoughthese agents did not induce acetylation of a-tubulin.These data are consistent with another study showingthat MS-275, an HDACi with specificity for HDACs 1,2, and 3 (3), can also induce Hsp90 acetylation anddegradation of the Flt3 client protein (52) and areinconsistent with a proposed mode of action of HDACiwhich posited that a specific inhibition of HDAC6 leadsto acetylation of HSP90 and disruption of its chaperonefunction, resulting in depletion of client proteins suchas Bcr-Abl and Flt-3 (25–28).

In summary, we have used MRLB-223 (inhibitsHDACs 1 and 2) as a novel tool compound to demon-strate that the inhibition of multiple HDACs by FDA-approved agents such as romidepsin (inhibits HDACs 1,2, and 3) and vorinostat (inhibits HDACs 1, 2, 3, 6, and 8)provided a superior antitumor response. Moreover, ourstudies allowed us to investigate the functional impor-tance of inhibition of HDAC6 in mediating the antitumoreffects of HDACis. It has been proposed that inhibition ofHDAC6 resulting in hyperacetylation of Hsp90 andrelease and degradation of client oncoproteins such asBcr-Abl is important for the antitumor activities of HDA-Cis (3, 10). However, our studies, using sophisticated"oncogene-addiction"models provide clear evidence thatinhibition of HDAC6 is not necessary for HDACis to killtumor cells that rely on the Hsp90 client protein Bcr-Ablfor survival. Moreover, we demonstrate that althoughdepletion of HDAC6 was sufficient to induce hyperace-tylation of HDAC6, this did not result in degradation ofBcr-Abl nor loss of viability of tumor cells "addicted" toBcr-Abl. This information is very important to the field as






it not only helps define the molecular events that arenecessary for HDACi-mediated antitumor responses,but provides a cautionary note regarding any proposeduse of acetylated Hsp90 as a biomarker for response toHDACis.

Disclosure of Potential Conflicts of InterestA. M. Kral, N. D. Ozerova, T. A. Miller, J. L. Methot, V. M. Richon, and

J. P. Secrist are current or former employees of Merck. No potentialconflicts of interest were disclosed by the other authors.

Authors' ContributionsConception and design: R.W. Johnstone, A. Newbold, G.M. Matthews, S.Chiocca, T.A. Miller, J.L. Methot, V.M. Richon, P. Secrist, S. MinucciDevelopment of methodology: A. Newbold, G.M. Matthews, L.A. Cluse,C.J.P. Clarke, C. Cullinane, J.E. Bolden, R.A. Dickins, S. ChioccaAcquisition of data (provided animals, acquired and managed patients,provided facilities, etc.): A. Newbold, G.M. Matthews, L.A. Cluse, K-M.Banks, C. Cullinane, A.J. Christiansen, A.M. Kral, N.D. Ozerova, S.MinucciAnalysis and interpretation of data (e.g., statistical analysis, biostatis-tics, computational analysis):R.W. Johnstone, A.Newbold,M. Bots, C.J.P.Clarke, J.E. Bolden, A.J. Christiansen, A.M. Kral, V.M. Richon, P. Secrist, S.MinucciWriting, review, and/or revision of the manuscript: A. Newbold, G.M.Matthews, C.J.P. Clarke, T.A. Miller, S. Minucci

Administrative, technical, or material support (i.e., reporting or orga-nizing data, constructing databases): A. Newbold, G.M. Matthews, C.MiccoloStudy supervision: R.W. Johnstone, A. Newbold, S. Minucci

AcknowledgmentsThe authors thankDonnaDorow, Ekaterina Bogatyreva, Susan Jackson,

LauraKirby,AgneseCollino, andAlfonsoPassafaro for technical help, andSalvatore Pece and Pier Paolo Di Fiore of theMolecularMedicine Programat the European Institute of Oncology for technical help and discussions.

Grant SupportR.W. Johnstone is a principal research fellowof theNationalHealth and

Medical Research Council of Australia (NHMRC) and supported by theNHMRC Program and Project Grants, Cancer Council Victoria, TheLeukemia Foundation of Australia, Victorian Breast Cancer ResearchConsortium, and Victorian Cancer Agency. R.A. Dickins is supported bythe NHMRC, Victorian Endowment for Science, Knowledge and Innova-tion (VESKI) (Australia), and The Sylvia and Charles Viertel Foundation.Work in the laboratory of S.Minucci has been supportedby theAssociationfor International Cancer Research (AIRC), Ministry of Health.

The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely to indicatethis fact.

ReceivedAugust6, 2013; acceptedAugust19, 2013; publishedOnlineFirstOctober 3, 2013.

References1. Johnstone RW. Histone-deacetylase inhibitors: novel drugs for the

treatment of cancer. Nat Rev Drug Discov 2002;1:287–99.2. Marks PA, Miller T, Richon VM. Histone deacetylases. Curr Opin

Pharmacol 2003;3:344–51.3. Bolden JE, Peart MJ, Johnstone RW. Anticancer activities of histone

deacetylase inhibitors. Nat Rev Drug Discov 2006;5:769–84.4. Kouzarides T. Acetylation: a regulatory modification to rival phos-

phorylation? EMBO J 2000;19:1176–9.5. Glozak MA, Sengupta N, Zhang X, Seto E. Acetylation and deacetyla-

tion of non-histone proteins. Gene 2005;363:15–23.6. Kim SC, Sprung R, Chen Y, Xu Y, Ball H, Pei J, et al. Substrate and

functional diversity of lysine acetylation revealed by a proteomicssurvey. Mol Cell 2006;23:607–18.

7. Choudhary C, Kumar C, Gnad F, Nielsen ML, Rehman M, Walther TC,et al. Lysine acetylation targets protein complexes and co-regulatesmajor cellular functions. Science 2009;325:834–40.

8. ZhouQ, Chaerkady R, ShawPG, Kensler TW, Pandey A, Davidson NE.Screening for therapeutic targets of vorinostat by SILAC-based prote-omic analysis in human breast cancer cells. Proteomics 2010;10:1029–39.

9. Spiegel S, Milstien S, Grant S. Endogenous modulators and pharma-cological inhibitors of histone deacetylases in cancer therapy. Onco-gene 2012;31:537–51.

10. Minucci S, Pelicci PG. Histone deacetylase inhibitors and the promiseof epigenetic (and more) treatments for cancer. Nat Rev Cancer2006;6:38–51.

11. Rasheed WK, Johnstone RW, Prince HM. Histone deacetylaseinhibitors in cancer therapy. Expert Opin Investig Drugs 2007;16:659–78.

12. Insinga A, Monestiroli S, Ronzoni S, Gelmetti V, Marchesi F, Viale A,et al. Inhibitors of histone deacetylases induce tumor-selective apo-ptosis through activation of the death receptor pathway. Nat Med2005;11:71–6.

13. Lindemann RK, Newbold A, Whitecross KF, Cluse LA, Frew AJ, Ellis L,et al. Analysis of the apoptotic and therapeutic activities of histonedeacetylase inhibitors by using a mouse model of B cell lymphoma.Proc Natl Acad Sci U S A 2007;104:8071–6.

14. Johnstone RW, Licht JD. Histone deacetylase inhibitors in cancertherapy: is transcription the primary target? Cancer Cell 2003;4:13–8.

15. BantscheffM, Hopf C, SavitskiMM, DittmannA, Grandi P,MichonAM,et al. Chemoproteomics profiling of HDAC inhibitors reveals selectivetargeting of HDAC complexes. Nat Biotechnol 2011;29:255–65.

16. Balasubramanian S, Verner E, Buggy JJ. Isoform-specific histonedeacetylase inhibitors: the next step? Cancer Lett 2009;280:211–21.

17. Haggarty SJ, Koeller KM, Wong JC, Grozinger CM, Schreiber SL.Domain-selective small-molecule inhibitor of histone deacetylase 6(HDAC6)-mediated tubulin deacetylation. Proc Natl Acad Sci U S A2003;100:4389–94.

18. Inoue S, Mai A, Dyer MJ, Cohen GM. Inhibition of histone deacetylaseclass I but not class II is critical for the sensitization of leukemic cells totumor necrosis factor-related apoptosis-inducing ligand-inducedapo-ptosis. Cancer Res 2006;66:6785–92.

19. Haberland M, Johnson A, Mokalled MH, Montgomery RL, Olson EN.Genetic dissection of histone deacetylase requirement in tumor cells.Proc Natl Acad Sci U S A 2009;106:7751–5.

20. Glaser KB, Li J, Staver MJ, Wei RQ, Albert DH, Davidsen SK. Role ofclass I and class II histone deacetylases in carcinoma cells usingsiRNA. Biochem Biophys Res Commun 2003;310:529–36.

21. Wilson AJ, Byun DS, Popova N, L'Italien K, Sowa Y, Arango D, et al.Histone Deacetylase 3 (HDAC3) and other Class I HDACs regulatecolon cell maturation and p21 expression and are deregulated inhuman colon cancer. J Biol Chem 2006;281:13548–58.

22. Whitesell L, Lindquist SL. HSP90 and the chaperoning of cancer. NatRev Cancer 2005;5:761–72.

23. Budillon A, Bruzzese F, Di Gennaro E, Caraglia M. Multiple-targetdrugs: inhibitors of heat shock protein 90 and of histone deacetylase.Curr Drug Targets 2005;6:337–51.

24. Kovacs JJ, Murphy PJ, Gaillard S, Zhao X, Wu JT, Nicchitta CV, et al.HDAC6 regulates Hsp90 acetylation and chaperone-dependent acti-vation of glucocorticoid receptor. Mol Cell 2005;18:601–7.

25. Fuino L, Bali P, Wittmann S, Donapaty S, Guo F, Yamaguchi H, et al.Histone deacetylase inhibitor LAQ824 down-regulates Her-2 andsensitizes human breast cancer cells to trastuzumab, taxotere, gem-citabine, and epothilone B. Mol Cancer Ther 2003;2:971–84.

26. Nimmanapalli R, Fuino L, Bali P, Gasparetto M, Glozak M, Tao J, et al.Histone deacetylase inhibitor LAQ824 both lowers expression andpromotes proteasomal degradation of Bcr-Abl and induces apoptosisof imatinib mesylate-sensitive or -refractory chronic myelogenousleukemia-blast crisis cells. Cancer Res 2003;63:5126–35.

Newbold et al.





27. Bali P,GeorgeP,CohenP, Tao J,GuoF,SiguaC, et al. Superior activityof the combination of histone deacetylase inhibitor LAQ824 and theFLT-3 kinase inhibitor PKC412 against human acute myelogenousleukemia cells with mutant FLT-3. Clin Cancer Res 2004;10:4991–7.

28. Yu X, Guo ZS, Marcu MG, Neckers L, Nguyen DM, Chen GA, et al.Modulation of p53, ErbB1, ErbB2, and Raf-1 expression in lung cancercells by depsipeptide FR901228. J Natl Cancer Inst 2002;94:504–13.

29. Bali P, Pranpat M, Bradner J, Balasis M, Fiskus W, Guo F, et al.Inhibition of histone deacetylase 6 acetylates and disrupts the chap-erone function of heat shock protein 90: a novel basis for antileukemiaactivity of histone deacetylase inhibitors. J Biol Chem 2005;280:26729–34.

30. Dexter TM,Garland J, Scott D, Scolnick E,Metcalf D. Growth of factor-dependent hemopoietic precursor cell lines. J Exp Med 1980;152:1036–47.

31. Vert JP, Foveau N, Lajaunie C, Vandenbrouck Y. An accurate andinterpretablemodel for siRNA efficacy prediction. BMCBioinformatics2006;7:520.

32. Peart MJ, Tainton KM, Ruefli AA, Dear AE, Sedelies KA, O'Reilly LA,et al. Novel mechanisms of apoptosis induced by histone deacetylaseinhibitors. Cancer Res 2003;63:4460–71.

33. Essmann F, Engels IH, Totzke G, Schulze-Osthoff K, Janicke RU.Apoptosis resistance of MCF-7 breast carcinoma cells to ionizingradiation is independent of p53 and cell cycle control but caused bythe lack of caspase-3 and a caffeine-inhibitable event. Cancer Res2004;64:7065–72.

34. Berk SC, Close J, Hamblett C, Heidebrecht RW, Kattar SD, Kliman LT,et al. inventors. Spirocyclic compounds. 7,544,695 U.S. Patent, 2009,June 9.

35. Ruefli AA, Ausserlechner MJ, Bernhard D, Sutton VR, Tainton KM,KoflerR, et al. The histone deacetylase inhibitor and chemotherapeuticagent suberoylanilide hydroxamic acid (SAHA) induces a cell-deathpathway characterized by cleavage of Bid and production of reactiveoxygen species. Proc Natl Acad Sci U S A 2001;98:10833–8.

36. Whitecross KF, Alsop AE, Cluse LA, Wiegmans A, Banks KM, Coo-mansC, et al. Defining the target specificity of ABT-737 and synergisticantitumor activities in combinationwith histone deacetylase inhibitors.Blood 2009;113:1982–91.

37. Hubbert C, Guardiola A, Shao R, Kawaguchi Y, Ito A, Nixon A, et al.HDAC6 is a microtubule-associated deacetylase. Nature 2002;417:455–8.

38. ZhangY, Li N,CaronC,MatthiasG, HessD, Khochbin S, et al. HDAC-6interacts with and deacetylates tubulin and microtubules in vivo.EMBO J 2003;22:1168–79.

39. GorreME,MohammedM, EllwoodK,HsuN, Paquette R, RaoPN, et al.Clinical resistance to STI-571 cancer therapy caused by BCR-ABLgene mutation or amplification. Science 2001;293:876–80.

40. Rao R, Nalluri S, Kolhe R, Yang Y, Fiskus W, Chen J, et al. Treatmentwith panobinostat induces glucose-regulated protein 78 acetylation

and endoplasmic reticulum stress in breast cancer cells. Mol CancerTher 2010;9:942–52.

41. Kekatpure VD, Dannenberg AJ, Subbaramaiah K. HDAC6 modulatesHsp90 chaperone activity and regulates activation of aryl hydrocarbonreceptor signaling. J Biol Chem 2009;284:7436–45.

42. Rasheed W, Bishton M, Johnstone RW, Prince HM. Histone deace-tylase inhibitors in lymphoma and solid malignancies. Expert RevAnticancer Ther 2008;8:413–32.

43. Vrana JA, Decker RH, Johnson CR, Wang Z, Jarvis WD, Richon VM,et al. Induction of apoptosis in U937 human leukemia cells by sub-eroylanilide hydroxamic acid (SAHA) proceeds through pathways thatare regulated by Bcl-2/Bcl-XL, c-Jun, and p21CIP1, but independentof p53. Oncogene 1999;18:7016–25.

44. Singh TR, Shankar S, Srivastava RK. HDAC inhibitors enhance theapoptosis-inducing potential of TRAIL in breast carcinoma. Oncogene2005;24:4609–23.

45. Bernhard D, Ausserlechner MJ, Tonko M, L€offler M, Hartmann BL,Csordas A, et al. Apoptosis induced by the histone deacetylaseinhibitor sodium butyrate in human leukemic lymphoblasts. FASEBJ 1999;13:1991–2001.

46. Rosato RR, Almenara JA, Dai Y, Grant S. Simultaneous activation ofthe intrinsic and extrinsic pathways by histone deacetylase (HDAC)inhibitors and tumor necrosis factor-related apoptosis-inducingligand (TRAIL) synergistically induces mitochondrial damage andapoptosis in human leukemia cells. Mol Cancer Ther 2003;2:1273–84.

47. Mitsiades N, Mitsiades CS, Richardson PG, McMullan C, Poulaki V,Fanourakis G, et al. Molecular sequelae of histone deacetylase inhi-bition in human malignant B cells. Blood 2003;101:4055–62.

48. Zhang XD, Gillespie SK, Borrow JM, Hersey P. The histone deacety-lase inhibitor suberic bishydroxamate regulates the expression ofmultiple apoptotic mediators and induces mitochondria-dependentapoptosis of melanoma cells. Mol Cancer Ther 2004;3:425–35.

49. Newbold A, Lindemann RK, Cluse LA, Whitecross KF, Dear AE,Johnstone RW. Characterisation of the novel apoptotic and therapeu-tic activities of the histone deacetylase inhibitor romidepsin. MolCancer Ther 2008;7:1066–79.

50. Ellis L, Bots M, Lindemann RK, Bolden JE, Newbold A, Cluse LA, et al.The histone deacetylase inhibitors LAQ824 and LBH589 do not requiredeath receptor signaling or a functional apoptosome tomediate tumorcell death or therapeutic efficacy. Blood 2009;114:380–93.

51. Bhaskara S, Chyla BJ, Amann JM, Knutson SK, Cortez D, Sun ZW,et al. Deletion of histone deacetylase 3 reveals critical roles in S phaseprogression and DNA damage control. Mol Cell 2008;30:61–72.

52. Nishioka C, Ikezoe T, Yang J, Takeuchi S, Koeffler HP, Yokoyama A.MS-275, a novel histone deacetylase inhibitor with selectivity againstHDAC1, induces degradation of FLT3 via inhibition of chaperonefunction of heat shock protein 90 in AML cells. Leuk Res 2008;32:1382–92.






2013;12:2709-2721. Published OnlineFirst October 3, 2013.Mol Cancer Ther Andrea Newbold, Geoffrey M. Matthews, Michael Bots, et al. with Diverse SpecificitiesMolecular and Biologic Analysis of Histone Deacetylase Inhibitors

Updated version

10.1158/1535-7163.MCT-13-0626doi:

Access the most recent version of this article at:

Material

Supplementary

http://mct.aacrjournals.org/content/suppl/2013/10/03/1535-7163.MCT-13-0626.DC1

Access the most recent supplemental material at:

Cited articles

http://mct.aacrjournals.org/content/12/12/2709.full#ref-list-1

This article cites 51 articles, 25 of which you can access for free at:

Citing articles

http://mct.aacrjournals.org/content/12/12/2709.full#related-urls

This article has been cited by 7 HighWire-hosted articles. Access the articles at:

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

Subscriptions

Reprints and

[email protected]

To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://mct.aacrjournals.org/content/12/12/2709To request permission to re-use all or part of this article, use this link



http://mct.aacrjournals.org/lookup/doi/10.1158/1535-7163.MCT-13-0626

http://mct.aacrjournals.org/content/suppl/2013/10/03/1535-7163.MCT-13-0626.DC1

http://mct.aacrjournals.org/content/12/12/2709.full#ref-list-1

http://mct.aacrjournals.org/content/12/12/2709.full#related-urls

http://mct.aacrjournals.org/cgi/alerts

mailto:[email protected]

http://mct.aacrjournals.org/content/12/12/2709


molecular and biologic analysis of histone deacetylase ... · victoria m. richon5, j. paul...

Documents