disulfiram and its novel derivative sensitize prostate ...download.xuebalib.com/37a8wskz3cxh.pdf ·...

MOLECULAR CARCINOGENESIS 55:1843–1857 (2016)

Disulfiram and Its Novel Derivative Sensitize ProstateCancer Cells to the Growth Regulatory Mechanisms ofthe Cell by Re-Expressing the Epigenetically RepressedTumor Suppressor—Estrogen Receptor bVikas Sharma,1 Vikas Verma,1 Nand Lal,2 Santosh K. Yadav,1 Saumya Sarkar,1 Dhanaraju Mandalapu,2

Konica Porwal,1 Tara Rawat,2 J.P. Maikhuri,1 Singh Rajender,1 V.L. Sharma,2 and Gopal Gupta1*1Division of Endocrinology, CSIR-Central Drug Research Institute, Lucknow, India2Division of Medicinal and Process Chemistry, CSIR-Central Drug Research Institute, Lucknow, India

Estrogen Receptor-b (ER-b), a tumor-suppressor in prostate cancer, is epigenetically repressed by hypermethylation ofits promoter. DNA-methyltransferases (DNMTs), which catalyze the transfer of methyl-groups to CpG islands of genepromoters, are overactive in cancers and can be inhibited by DNMT-inhibitors to re-express the tumor suppressors. TheFDA-approved nucleoside DNMT-inhibitors like 5-Azacytidine and 5-Aza-deoxycytidine carry notable concerns due totheir off-target toxicity, therefore non-nucleoside DNMT inhibitors are desirable for prolonged epigenetic therapy.Disulfiram (DSF), an antabuse drug, inhibits DNMT and prevents proliferation of cells in prostate and other cancers,plausibly through the re-expression of tumor suppressors like ER-b. To increase the DNMT-inhibitory activity of DSF, itschemical scaffold was optimized and compound-339 was discovered as a doubly potent DSF-derivative with similar off-target toxicity. It potently and selectively inhibited cell proliferation of prostate cancer (PC3/DU145) cells in comparison tonormal (non-cancer) cells by promoting cell-cycle arrest and apoptosis, accompanied with inhibition of total DNMTactivity, and re-expression of ER-b (mRNA/protein). Bisulfite-sequencing of ER-b promoter revealed that compound-339demethylated CpG sites more efficaciously than DSF, restoring near-normal methylation status of ER-b promoter.Compound-339 docked on to the MTase domain of DNMT1 with half the energy of DSF. In xenograft mice-model, thetumor volume regressed by 24%and 50%after treatmentwith DSF and compound-339, respectively, with increase in ER-b expression. Apparently both compounds inhibit prostate cancer cell proliferation by re-expressing the epigeneticallyrepressed tumor-suppressor ER-b through inhibition of DNMT activity. Compound-339 presents a new lead for furtherstudy as an anti-prostate cancer agent. © 2015 Wiley Periodicals, Inc.

Key words: disulfiram; estrogen receptor-b; DNMT1; prostate cancer; cell proliferation

INTRODUCTION

Prostatic cancer (PCa), a leading cause of cancer-related death inmen over 60, is now being consideredas a severe and challenging medical problem world-wide [1]. PCa is characterized by intricate genomicalterations that are highly diverse and alter im-mensely from patient to patient [2]. PCa has alsobeen considered as an “epigenome disaster” sinceaberrations in DNA methylation, histone modifica-tions, and miRNA patterns have been detected wellbefore the cancer becoming invasive, and persist tillthe end [3]. Hence, epigenetic changes are imminentin prostate cancer [4].DNA methylation, a key epigenetic modification

involved in gene regulation is reversible and occursearly in human PCa and other cancers [5–10],making it an attractive target in cancer therapeu-tics [8]. Three DNA methyltransferases (DNMTs),DNMT1, DNMT3a, and DNMT3b, catalyze thetransfer of a methyl group from the donor moleculeS-adenosyl-methionine (SAM) to a cytosine in theCpG islands of gene promoter, resulting in gene

silencing [11–13]. About 30 genes have been knownto be epigenetically silenced through hyper-meth-ylation [10], including the estrogen receptor-beta(ER-b), which has a growth inhibitory role inprostatic tissue and is a proposed therapeutic targetfor prostate cancer [14]. ER-b contains a typical CpGisland within its promoter [15] and methylation-specific PCR studies have shown that the promoterregion of the ER-b gene is hypermethylated inprostate cancer tissues [16–18]. Inhibition of DNAmethylation might restore susceptibility of PCa cells

Grant sponsor: Ministry of Health and Family Welfare (CDRIcommunication number 9121)*Correspondence to: Division of Endocrinology, CSIR-Central Drug

Research Institute, Sector-10, Jankipuram Extension, Sitapur Road,Lucknow 226031.Received 16 August 2015; Revised 20 October 2015; Accepted 3

November 2015DOI 10.1002/mc.22433Published online 24 November 2015 in Wiley Online Library

(wileyonlinelibrary.com).

� 2015 WILEY PERIODICALS, INC.

to normal growth-regulatory mechanisms withincell, and the AR-deficient hormone refractoryprostate cancer (HRPC) cells to ER-b agonists. TwoFDA approved nucleoside DNMT inhibitors 5-azaCand 5-aza-dC are being investigated clinically in PCaand other solid malignancies. However, these drugscarry considerable concerns regarding carcinogene-sis since they are incorporated into the DNA duringS-phase and inhibit DNMT by irreversibly trappingthe enzyme to DNA, risking DNA strand breaks andmutations, which can cause severe side-effects andtoxicity in long term use [19]. Therefore, non-nucleoside DNMT inhibitors such as Disulfiram(DSF) [20] may prove safer in prolonged epigenetictherapy. DSF is a drug in clinical use for themanagement of chronic alcoholism as it containsstrong thiol-reactive functional groups which inter-act with the thiol group of reactive cysteine at theactive site of the alcohol-metabolizing enzymealdehyde dehydrogenase [21]. It is prescribed at adaily dose of 500mg for up to 6 months, whichindicates its high levels of safety in normal subjects.Incidentally, DNMT1 also contains a reactiveCXXC region (C is cysteine; X is any other aminoacid) at its active site, which is essential forenzymatic activity [22] and makes it susceptible toDSF [20]. Since inhibition of DNMT function canpotentially reverse some of the cancer-associatedmethylation marks [23] and lead to reprogrammingof the epigenetic make up of cancer cells, itrepresents an attractive therapeutic avenue [8]. Wetherefore hypothesized that DSF could reactivate thetumor suppressor ER-b in PCa cells by its DNMTinhibitory activity and re-sensitize them to growthregulatory mechanisms of the cell. After establishingDSF activity against PCa cells in preliminary experi-ments, we took up synthesis of a series of DSFderivatives, to increase the activity of the molecule.Chemical groups were introduced in DSF to increaseits lipophilicity (for increasing bioavailability) andto maintain (if not increase) the electro-negativity ofits disulphide group (for interaction with thiols).The present study highlights the potential of DSFand one of its most promising novel derivatives tore-express the epigenetically repressed ER-b gene inPCa cell lines by inhibiting DNMT activity andthereby sensitizing cancer cells to apoptotic and cellcycle regulatory mechanisms of the cell.

MATERIALS AND METHODS

Cell Cultures

The prostate cancer PC3, monkey kidney Cos-7,and human embryonic kidney-293 (HEK293) celllines were procured from National Centre for CellScience, Pune, India. Prostate cancer DU145 andnormal prostate epithelial RWPE-1 cell lines werefrom ATCC (Rockville, MD). Human prostate cancercells, PC3 and DU145, were maintained in DMEM/

HAM’S F-12 medium (Sigma–Aldrich, St. Louis, MO),and non-prostatic, non-cancer, normal cell lines Cos-7 andHEK293were grown inDMEMmedium (Sigma–Aldrich, St. Louis, MO). Both the media weresupplemented with 10% FBS and 0.01% antibiotic/antimycotic solutions. Normal prostate cell lineRWPE-1, was maintained in KSFM medium (Invitro-gen, Carlsbad, CA) supplemented with epidermalgrowth factor and bovine pituitary extract. All cellswere maintained at 378C in 5% CO2 atmosphere.

MTT Assay for Cell Viability

Prostate cancer and normal cells were seeded in96-well plates and incubated in culture mediumcontaining 10% FBS and antibiotic/antimycoticsolution. At 50% confluency, cells were serumdeprived for 6 h and then treated with differentconcentrations of DSF or its derivative compound-339 for 24 and 48h. Thereafter MTT solution (5mg/ml in PBS, pH 7.4) was added to each well. Theformazan crystals formed inside the viable cells weresolubilized in DMSO and the absorbance wasmeasured at 540nm using a microplate reader(Microquant, Bio-tek, Winooski, VT). The effective/inhibitory concentration-50 (EC50/IC50) were calcu-lated by using Prism3.0 software.

Cell Cycle Analysis by Flow Cytometry

Subsequent to incubation with different concen-trations of test compounds for 48h, the PC3 andDU145 cells were washed with PBS, harvested andfixed in 75% ethanol overnight at 48C. Thereafter,cells were washed with PBS, resuspended in DNAextraction buffer and incubated at room temperaturewith RNase A (20mg/ml) for 15min. The cell pelletwas stained with 30.0mg/ml propidium iodide (PI) for15min, washed, and analyzed on a Flow Cytometer(FACS Calibur, BD Biosciences, Franklin Lakes, NJ).The percentage of cells in different phases of cell cyclewas analyzed by ModFit LT 3.0 software (VeritySoftware House).

Flow Cytometric Evaluation of Cell Apoptosis

Apoptosis induced in PC3 and DU145 cells by thetest compounds was quantitatively measured byAnnexin V-FITC/PI assay. Cells, after 48h of treat-ment with compounds or vehicle, were harvested bytrypsinisation and collected by centrifugation, andwashed twice with PBS. The cell pellets were resus-pended in 1� binding buffer. After the addition ofAnnexin V-FITC and PI (Propidium Iodide) to thebuffer, the tubes were incubated in the dark. Thepercentage of cells positive for Annexin V-FITC and PIwas determined in a flow cytometer (Model FACSCalibur, BD Biosciences, Franklin Lakes, NJ) equippedwith an argon laser (488nm) for excitation, using thethreshold signals for normal cells. This assay wascarried out following manufacturer’s instructionsprovided on kit by Roche (Basel, Switzerland).

1844 SHARMA ET AL.

Molecular Carcinogenesis

DNMT Activity

Cells

PC3 and DU145 cells treated with either DSF (1mM,2mM,and3mM)or compound-339 (0.5mM,1mM,and2mM) for 48h were harvested and nuclear extractswere prepared using nuclear extraction kit (ActiveMotif, Carlsbad, CA) as per the manufacturer’s proto-col. The DNMT activity was measured using 10mg ofnuclear extract, after an incubation period of 2h, withthe help of a DNA methyltransferase activity assay kit(Active Motif, Carlsbad, CA) by following the manu-facturer’s protocol. Results were expressed as percentDNMT activity compared to the control (100%).Five mM of 5-Azacytidine was used as positive control.

Tissues

Total DNMT activity was also measured in tumortissues from xenograft mice model. Fixed amount oftissues from different groups were weighed and thetissue extracts were prepared according to the kitmanufacturer’s protocol. DNMT activity was mea-sured using the DNA methyltransferase activity assaykit (Active Motif, Carlsbad, CA).

Global DNA Methylation Quantification

Methylated CpGs were quantified from genomicDNA to determine relative methylation states amongdifferent DNA samples. PC3 and DU145 cells wereallowed to grow and then treated with differentconcentrations ofDSF or its derivative compound-339or vehicle (control) for 48h. Cells were thenharvestedand their genomic DNA was extracted using a DNAextraction kit (Invitrogen, Carlsbad, CA). Relativeglobal methylation levels were measured usingmethylated DNA quantification kit (MDQ1, Sigma-Aldrich, St. Louis, MO). A known DNMT inhibitor 5-Azacytidine (5mM) was used as positive control.

mRNA Expression by Real Time RT-PCR

Total RNA was isolated using Trizol reagent (Invi-trogen Life technologies, Carlsbad, CA) and 3mg ofRNA was converted to cDNA using the RevertAid HMinus First Strand cDNA Synthesis Kit (Fermentas,Waltham, MA) following the manufacturer’s instruc-tions. Real time PCR was performed on Light Cycler480 (Roche, Basel, Switzerland) detection systemusingSYBRGreen IMastermix (Roche, Basel, Switzerland) in96-well plates. All reactions were run in triplicate andrelative gene expression was normalized to GAPDH,calculations made by using the 2-DDCt method.

Protein Expression by Western Blot

Total cell lysates were prepared in cell lytic solution(Sigma-Aldrich, St. Louis, MO). The cell lysis solutionwas supplemented with protease inhibitor cocktail(Sigma-Aldrich, St. Louis, MO) and phosphataseinhibitor cocktail (Sigma-Aldrich, St. Louis, MO).Cell lysates containing equal amount of protein, asestimated by using Bradford reagent (Sigma-Aldrich),were subjected to immunoblotting. Immunoblottingwas performed using following antibodies: estrogenreceptor-b, estrogen receptor-a (Santacruz Biotech,Dallas, TX), Bax, Bcl2, PCNA, PS2, TNF-a, cell cycleinhibitors (P21 and P27), pAKT and total AKT, pJNK,and total JNK (Cell Signaling Technology, Beverly,MA).

ER-b Protein Expression by Immunocytochemistry (ICC)

PC3 cells were treated with either DSF (2mM) orcompound-339 (1mM) or estradiol or compoundswith estradiol, and were incubated for 48h. Thereaf-ter, cells were fixed in 4% paraformaldehyde andpermeabilized in 0.25% Triton X-100 in PBS. Cellswere incubated with 1% BSA diluted in PBST for30min with gentle rocking to block non-specificbinding. Finally, cells were immunostained withdesired concentration of primary antibody (ER-b[Santacruz Biotech, Dallas, TX]) and secondary anti-body (Anti-goat IgG-Texas Red [Santacruz Biotech,Dallas, TX]) in PBST with 1% BSA. Cells were thenmounted with Prolong gold antifade reagent withDAPI (Invitrogen, Carlsbad, CA) and visualized undera microscope (Nikon 80i, Nikon Corporation, Tokyo,Japan). Imageswere captured usingNIS-elements F3.0software.

Reporter Gene Assay by Luciferase Expression

PC3 cells were seeded at a density of 5�104 cells/well into 24-well plates on the day prior to transfec-tion with 1mg of luciferase reporter gene constructpSG5 2xERE-Luc using DharmaFECT transfectionreagent and plus-transfection reagent (Invitrogen,Carlsbad, CA) according to maufacturer’s protocol.After transfection, cells were treated with test com-pounds per se or in combination with ER-b agonist(DPN) and ER-a agonist (PPT) for 48h. Luciferaseactivity was determined with luciferase assay systems(Promega, Madison, WI) following the manufac-turer’s protocol to detect the ER-ERE mediatedtranscriptional activity. Luciferase activity was nor-malized for transfection efficiency using pRL-SV40-lucas an internal control.

Small Interfering (si)RNA Mediated Knockdown of ER-b

PC3 and DU145 cells were plated into 6-well plateand allowed to adhere for 24h. Firstly, scrambledsiRNA (random sequence) and ER-b siRNA wereincubated with liposomal based DharmaFECT trans-fection reagent, thereafter, the cells were transfected

Gene Primer sequence (50-30)Estrogen

Receptor-a

Forward-GCATAAGAAGACAGTCTCTGAGTGAT

Reverse-GTCCCGGAGAATGTGAAGAG

Estrogen

Receptor-b

Forward-GCATTCTACAGGCCAAATTCAGATAA

Reverse-GCCATACTTCCCTTGTCATTGGT

GAPDH Forward-TCATCCCTGCCTCTACTG

Reverse-TGCTTCACCACCTTCTTG

DISULFIRAM AND ITS NOVEL DERIVATIVE 1845


with either scrambled siRNA or ER-b siRNA in serumfree medium, incubated with complete medium for24h, and then the cells were switched to mediumwith or without compounds for 48h before analysis.After completion of treatment, cells observed under alight microscopy and thereafter their protein wascollected for western blot.

In Vivo Efficacy in Xenograft Mice Model

All experimental procedures related to animals wereconducted byACDSF at ACTREC,NaviMumbai, India.PC3 cells (2�107) were injected subcutaneously intothe right flank area of male NOD-SCID mice aged 4–6wk. When the estimated tumor volume was reached(�100mm3), mice were randomly assigned into threegroups (five mice per group): Control (vehicle), DSF,and compound-339. Treatment group mice receivedDSF (10mg/kg, orally) or compound-339 (10mg/kgorally) every day for 18d. Bodyweightwasmeasuredatregular time intervals. Mice were sacrificed on theday following the completion of treatment and theinitial and final tumor volumes were calculated usingthe formula: Tumor Volume (cc)¼ [(d1þd2)/2)

3]�0.5236)/1000. [(d1þd2)/2]¼ the average diameterwhich is raised to the power 3 (Figure 1B). At autopsy,tumors and prostate glands were removed and dividedinto two pieces, one ofwhichwas snap frozen in liquidnitrogen, and the other was fixed in 10% bufferedformalin, embedded inparaffin for further histologicalanalysis.

Hematoxylin and Eosin (H&E) and Immunohistochemical(IHC) Staining

Tumor tissues from control and experimentalanimals were collected and processed for histologicalevaluations. Briefly, tissueswere fixed in 10%bufferedformalin, embedded in paraffin wax and 5mmsections were cut using a microtome (Leica Biosys-tems, Nussloch, Germany). Sections were processedfor H&E and IHC staining for ER-b and Ki-67 proteinsusing a kit (DAB150, Merck Millipore, Darmstadt,Germany), and analyzed under a light microscope(Nikon) and images were captured using software NISelements, at different magnifications.

Docking Studies

The 3D structure of DNMT1 was extracted fromProtein Data Bank (PDB id: 3PTA) [24]. Prior dockingsimulations, the DNMT1 structure was edited byadding missing residues and by removing heteroa-toms, including water molecules. Further, to obtainstable molecule energy, minimization was performedwith GROMOS96, an implementation of Swiss-PdbViewer version 4 [25]. The structures of DSF andcompound-339 were obtained from PubChemwith PubChem id CID 3117 and CID 14906453,respectively.Docking with Autodock-4.2 was performed fol-

lowing ADT routines [26]. Grid map was calculatedby AutoGrid with dimensions of 64A�64A�64A

Figure 1. Percent cell viability byMTT assay of PC3 (A) andDU145 (B)in presence of Disulfiram (DSF) and compound-339 at 24 h and 48h.Effective Concentration-50 (EC50) and EC25 of DSF and compound-339against prostate cancer PC3 and DU145 cells, Inhibitory

Concentration-50 (IC50) against human embryonic kidney (HEK) andnormal prostate epithelial (RWPE-1) cell lines and the selectivity index(SI) of test compounds (C).

1846 SHARMA ET AL.


and spacing of 0.375A between the grid points sothat MTase domain of DNMT1 can be covered.Further, the docking simulations were performedwith Lamarckian Genetic Algorithm (LGA), the mostefficient and reliable method of Autodock. Eachdocking experiment was performed for 100 runscontaining population size of 300 individuals and itwas terminated with maximum number of 250 0000energy evaluations or a maximum number of 27 000generations. Further for each run, the top rankedindividual in the population was set to survive intothe next generation. Mutation and crossover rateswere set at 0.02 and 0.80 respectively. Visualizationwas performed using UCSF Chimera1.6.2rc [27].

Bisulfite Sequencing

Methylation status of the 50 regulatory region of ER-b gene was analyzed by bisulfite sequencing. Bisulfiteconversion of genomic DNA (1mg) from PC3 andRWPE-1 cell lines was performed using the Methyl-CodeTM bisulfite conversion kit (Invitrogen, Carlsbad,CA) by following themanufacturer’s instructions. Thetarget region of converted DNA was amplified usingforward (5-GTATTTTTGTATGTGGTTAGAAATT-3)and reverse (5-TCCCCAAAACTAAAAAACATCC-3)primers. PCR conditions consisted of denaturationat 958C for 5min, followed by 35 cycles of amplifica-tion using 958C for 30 s, 588C for 45 s, 728C for 40 s,and a final stage at 728C for 7min. PCR products weregel purified and directly sequenced using Sangerchain termination method on a 3730 DNA analyzer(Applied Biosystems, Foster City, CA). Sequencecomparison was performed using Sequence Analysissoftware v5.2 (Applied Biosystems).

Statistical Analysis

All experiments were analysed by using a one-wayanalysis of variance (ANOVA) and post hoc Bonferro-ni’s test. Data and statistics were analyzed usingGraphPad Prism version 3.0 software. Graphicalresults represent mean� S.E. Experiments were donewith n�3.

RESULTS

Effect of Compounds on Cell Viability

The cell viabilities of prostate cancer cell lines (PC3and DU145) and normal cell lines (HEK293 andRWPE-1) treated with DSF or compound-339 (0mM–40mM) for 24 and 48h were assessed by MTT assay.Both the cell lines showed dose-dependent suscepti-bility (up to 20mM) to the treatments (Figure 1A andB). Figure 1C indicates that the EC50 (50% EffectiveConcentration) of DSF and compound-339 againstPC3 was 5mM and 2.5mM, respectively, and againstDU145 it was 5.3mM and 2.5mM. The safety/selectivity of action was almost 20-fold in case DSFand 40-fold in case of compound-339 as indicated bytheir IC50 (50% Inhibitory Concentration) against

non-prostatic normal cell lineHEK293 (DSF¼103mMand compound-339¼97mM). Similarly a 15-fold and30-fold selective action was against normal prostaticcell line RWPE-1 (DSF and compound-339¼75mM).Accordingly, compound-339 exhibited approxi-mately twofold higher Selectivity Index (TI) in vitro,than DSF.

Compounds Inhibit Cell Cycle Progression

Inhibition of prostate cancer cells viability bycompounds DSF and compound-339 was accompa-nied by distinct cell cycle arrest at the S-phase and aminor arrest at G2/M phase. The effect of DSF on PC3cell-cycle arrest was not statistically significant,whereas in case of compound-339, the effect of higherconcentration (2mM)was quite significant (P<0.001).Similarly in DU145 cell line, the higher dose ofcompound-339 (2mM) caused significant (P<0.01)cell cycle arrest at S-phase (Figure 2A).

Compounds Induce Apoptosis

Cell cycle arrest may lead to apoptosis. UsingAnnexinV-FITC/PI staining and flow cytometry, PC3and DU145 cell lines treated with different concen-trations of either DSF, or compound-339 or staur-osporine (positive control) showed that in 48h, thepercentage of early apoptotic cells was increasedfrom 5.87% (in control) to 22.93% with 1mMcompound-339 in PC3 cell line. However, a similarincrease of early apoptotic PC3 cells (�21.36%) wasseen with 3mM of DSF. Similarly in DU145 cellline, early apoptotic cells increased from 9.12% incontrol to 44.87% in compound-339 (2mM) treatedcells, and 36.93% in DSF (3mM) treated cells(Figure 2B).

Inhibition of Total DNMT Activity

The total DNMTactivity/inhibition assay innuclearextracts of PC3 cells revealed that compound-339treated extracts at 1mM and 2mM concentrationsshowed 34% and 49% inhibition of DNMT activity,respectively. Similarly in case of DU145 cells, a dosedependent inhibition of DNMT activity (47.4%,58.1%, and 74.7%) by compound-339 was observedat 0.5mM, 1mM, and 2mM concentrations, respec-tively. On the other hand, DSF inhibited DNMTactivity significantly only at 3.0mM in PC3 cells,whereas its inhibitory action in DU145 was notexactly dose dependent with significant inhibitionsat 1.0mM (P<0.01) and 3.0mM (P<0.05) (Figure 3A).However, no significant changes in the protein levelsof DNMTs were observed after treatment with bothcompounds (data not shown).

Compounds Decrease DNA Methylation Globally

Significant decrease in global DNAmethylationwasobserved in PC3 cell linewith 3mMDSF (P<0.01) and2mM compound-339 (P<0.05), at 48h. In case ofDU145, only compound-339 inhibited global DNA



methylation at 2mM (P<0.05). 5-Azacytidine inhib-ited global DNA methylation significantly at 5mM inboth cell lines (P<0.01)(Figure 3B).

Docking Studies

Before docking compound-339, we first tried toexplore the binding pocket for DSF (which is thebackbone of compound-339) in DNMT1 structure.The docking parameters were first verified by dockingthe S-adenosylhomocysteine (SAH) as a referenceligand to the MTase domain of DNMT1. Dockingresults showed 1.74 A RMSD values between crystal-lographic structure and predicted conformations ofDNMT1, thus verifying the correctness of parametersselected for docking simulations. The binding ener-gies (binding affinity) of compound-399 (�8.35 kcal/mol) was �twofold lower than that of DSF (�4.79kcal/mol) (Figure 4B). Both DSF and compound-339bound to the MTase domain comprising of residuesPRO1125, PHE1145, SER1152, GLU1168, MET1169,

TRP1170 (Figure 4A) to inhibit the interactionsbetween CXXC motif, which is required for catalyticactivity, of auto-inhibitory linker with DNA.

Time-and Dose-Dependent ER-b Re-Expression byCompounds

At EC25(1mM) of compound-339, the expression ofER-b and ER-a proteins in PC3 cell line, and ER-bprotein in DU145 cells was investigated at varioustime points. Estrogen receptor negative cell line (Cos-7) was used as negative control. In case of PC3 cellline, maximum expression of ER-bwas observed at 24and 48h (P<0.01) after treatment, with no signifi-cant change in ER-a expression as compared with 0h.Similarly, in DU145 cells, the expression of ER-bprotein was significantly increased (P<0.001) at 48and 72h, as compared with 0h (Figure 5A). Accord-ingly, for all further studies, cells were incubatedwith compounds for 48h. The protein expression ofER-b target gene TNF-awas also increasedwhereas the

Figure 2. (A) Cell cycle analysis by Flow Cytometry of PC3 andDU145 cells treated with disulfiram (DSF) (1mM, 2mM, or 3mM) andcompound-339 (0.5mM, 1mM, or 2mM) for 48 h. CCCP (3mM) wasused as positive control. Mean� SE of three independent experiments.(B) Externalization of phosphatidylserine and membrane permeabi-lization assay for apoptosis/necrosis of PC3 and DU145 cells by

Annexin-FITC/PI staining and Flow Cytometry. Disulfiram (DSF) andcompound-339 used at indicated concentrations for 48 h, staurospor-ine (3mM for 16 h) was used as positive control. Lower Left (LL)quadrant: viable cells; LR: early apoptotic cells; UR: late apoptotic cells;and UL: dead/necrotic cells. Mean� SE of three independentexperiments.

1848 SHARMA ET AL.


ER-a target gene PS2 remained unchanged in DSFand compound-339 treated PC3 cells. A similar dose-dependent increase in ER-b target TNF-a in DU145cell line at 48h was also observed (Figure 5B).

Effects of Compounds on the Estrogen Receptor-b (ER-b)Protein Expression (Immunocytochemistry)

In PC3 cells treated with DSF at 2mM andcompound-339 at 1mM for 36h, and then incubatedwith estradiol (E2) for 12h, ER-b (red)was translocated

into nucleus due to the presence of E2 (100nM).Interestingly, increased expression of ER-b wasobserved in presence of DSF and compound-339.However, the effect of compound-339 was morepronounced than DSF (Figure 6A).

Compounds Enhanced Transcriptional Activity of EREby ER-b

Increased expression of ER-b protein in compound-treated PC3 cells prompted us to determine the status

Figure 3. (A) Total DNMT activity in prostate cancer PC3 and DU145cell lines treated with disulfiram (DSF) and compound-339 for 48 h.Unit of DNMT activity¼OD/mg protein/hr. Control¼ 100; 5mM of 5-Azacytidine used as positive control. �P< 0.05, ��P< 0.01, ��P< 0.001. Means� SEM of two independent experiments in triplicates.

(B) Global DNA methylation (5-methyl cytosine content) of genomicDNA from PC3 and DU145 cell lines in presence of disulfiram (DSF) andcompound-339 at 48 h. Control¼ 100%. 5-Azacytidine (5mM) wasused as positive control.�P< 0.05, ��P< 0.01.

Figure 4. (A) Docking of disulfiram (DSF) and compound-339 in the Mtase Domain of DNMT1 along withS-adenosylhomocysteine (SAH). SAH: blue, DSF: red, compound-339: green. (B) The binding energies of thecompounds (kcal/mol).



of ER-b signalling in these cells. The increased estrogenreceptor signaling in treated cells was quantified by areporter gene assay. In PC3 cells that were transfectedwith ERE-Luciferase reporter plasmid, compound-339significantly increased luciferase activity at 1mM(P<0.05) in 48h. Interestingly, in the presence ofER-b agonist DPN (100nM), bothDSF and compound-339 further increased transcriptional activity by �2.5-fold (P<0.01) compared to ERE control. On the otherhand, ER-b antagonist THC repressed the compoundinducedER-b signalling (Figure 6B). It is thus indicatedthat compounds increase ER-b mediated signalling inprostate cancer cells.

Test Compounds and ER-b Silencing

In PC3 cells, ER-b protein levels increased signifi-cantly after treatment with compound-339 at 1.0mM(P<0.01) or DSF at 2.0mM (P<0.05). However,transfection of these cells with ERb siRNA broughtdown the ER-b protein to basal levels in DSF-treatedcells. On the other hand, compound-339 treatedcells maintained significantly higher ER-b proteinlevel in presence of siRNA as compared with control(P<0.05). Similarly, significant reduction (P<0.001)of ER-b protein in DU145 cells transfected with siRNAwas countered by the presence of compounds, andwhile compound-339 at 1.0mM increased ER-b tocontrol levels, DSF at 2.0mM increased it further(P<0.05). ER-b siRNA could knockdown �40%mRNA (P<0.01) (Figure 7A).

When observed under a microscope, ER-b silencedcells proliferated profusely (Supplementary Figure S2)and there was an upregulation of its proliferating cell

nuclear antigen (PCNA) and B-cell lymphoma associ-ated protein 2 (Bcl2) along with downregulation ofBax, as indicated by western blots of both the celllines. Treatment of these cells with the compoundscountered these effects of siRNA(Figure 7B).

Effect of ER-b Agonist (DPN) and Antagonist (THC) inPresence of ER-b Supplementing Compounds on CellViability of PC3 and DU145 Cells

DSF at 2mM, compound-339 at 1mM, and DPN at100nM significantly decreased (P<0.01) viability ofPC3 and DU145 cells, whereas THC (5mM) increasedit apparently. The effect of DPN was enhanced bypresence of DSF and compound-339 (P<0.001),whereas THC reduced inhibitory effects of com-pounds (Figure 8A). In parallel, the effect of ERsubtypes on cell viability was investigated in PC3 cellstreated with a fixed concentration of DSF andcompound-339. ER-a antagonist MPP (7.0mM) in-duced reduction of cell viability was supplemented bycompound-339 (P<0.001) and DSF (P<0.01). THCand MPP negated each other’s effect on cell viability(Figure 8B).

Methylation Status of ER-b in RWPE-1 and PC3 Cell Lines

The sequence of ER-b promoter region was obtainedfrom the Ensemble database. Two 50 CpG islandswithin the 50-regulatory region of ER-b promoterwere identified using MethPrimer (SupplementaryFigure S5). The region was located approximately250bp upstream of the transcription start site andincluded 27 CpG duplets. A high efficiency of DNAconversion with good quality DNA was obtained. In

Figure 5. (A) Time (0–72h) dependent ER-b and ER-a proteinsexpression at EC25 (1mM) of compound-339 in PC3 cell line and ER-bprotein expression in DU145 cell line (ER-a negative cells), ERa/b-negative [ER(�)] Cos-7 cell line was used as negative control. (B)

Concentration dependent expression of ER-b and ER-a proteins (inPC3) and ER-b (in DU145), and their downstream proteins TNF-a andPS2, respectively. �P< 0.05, ��P< 0.01 and ��P< 0.001.

1850 SHARMA ET AL.


the ER-b promoter region of normal prostate epithelialRWPE-1 cell line, only two out of 27 duplets (<10%)were methylated, whereas in prostate cancer PC3 cellline, 12 out of 27 CpG sites (45%) were methylated.While one of theCpG siteswas completelymethylated(100%), 11 other CpGs were methylated by �20–80%in PC3 cells. However, treatment of PC3 cells with DSFresulted in a marked reduction of the methylationstatus of almost all the 12 sites while treatment withcompound-339 completely abolished themethylationof several siteswith just 10%methylation at threeCpGsites and 20–30% methylation at four sites remainingafter treatment (Figure 9).

DSF and Compound-339 Reduce Tumor Growth inXenograft Mice Model

The Relative Tumor Volume (RTV), i.e. tumorvolume on day of measurement/tumor volume on

day 1, was plotted versus days of treatment. There wasa time dependent linear increase in tumor volume ofcontrol (vehicle-treated) mice with xenograft, andafter 18d of experiment, there was an approximatelysixfold increase. However in DSF treated group, after9th day of treatment, tumor volume regressedgradually and after 18 d of experiment, the volumeswere 24% lower than control. On the other hand, incompound-339 treated mice, the tumor volumeregressed from the 1st day of treatment and at theend of 18d treatment, the tumor volume was �50%lower than control (Figure 10A). There was nosignificant change in body weight (Figure 10B).

DNMT Activity of Tumor Tissues

In spite of significant difference in tumor volumesof mice treated with the two compounds, therewas �66% reduction in DNMT activity of tumor

Figure 6. (A) Representative images of untreated (control), estradiol(E2, 100 nM), E2 þdisulfiram (DSF) (100 nMþ 2.0mM), E2þ com-pound-339 (100nMþ 1.0mM) treated PC3 cells at 48 h. Fluorescence:ER-b(red), Nuclei (DAPI, blue). (B) Transactivation of ERE-Luc by ER-b inPC3 cell line was measured by dual-luciferase reporter assay in form of

RLU (Relative Light Unit). DSF (2mM), compound-339 (1mM) per se ortheir combination with ER-b agonist DPN (100 nM) or ER-b antagonistTHC (5mM) were assayed at 24 h. Mean� SD of three experiments.�P< 0.05, ��P< 0.01 versus ERE-Luc.



tissues obtained from animals of both the groups incomparisonwith control animals tissues (Figure 10C).

Hematoxylin and Eosin (H&E) and Immunohistochemical(IHC) Staining

H&E staining indicated reduced infiltration ofmacrophages, fewer tumor cells with vacuolizationor apoptosis in tumor tissues of treated animals ascompared with that of vehicle control, which con-sisted of tightly packed tumor cells. The expression ofKi-67 (proliferation marker) was significantly lower inDSF and compouind-339 treated mice tumors than incontrol. The tumor suppressor ER-b exhibited higherexpression in compound-treated mice tissues ascomparedwith control group.However, no significantdifference in ER-b protein was observed between DSFand compound-339 group animal tissues (Figure 10D).

DISCUSSION

Reverting epigenetic modifications associated withtumor progression has been considered as a valuabletherapeutic approach [8,13,28]. Taking basis of that,this first study clearly demonstrates that DSF and itsmore potent novel derivative compound-339 inhibitDNMTs (that play an important role in oncogenesis),causingCpGdemethylation and restoration of at least

one tumor suppressor gene (ER-b) in prostate cancercell lines PC3 and DU145.While DSF was 15 to 20 times selective for

inhibition of cancer cell proliferation, compound-339 was twofold more selective than DSF, withcomparable safety in vitro. Thus, meticulous optimi-zation of DSF scaffold yielded a novel structure(Compound-339) that was twofold more active thanits parent molecule (DSF), without any escalation ingeneral cell toxicity. A previous report has identifiedDSF as a selective inhibitor of prostate cancer cellsin vitro and in vivo [29]. However, recent reports havealso shown its anti-tumor effects against various othercancer cell lines including melanoma, leukaemia,small lung cell cancer, osteosarcoma, cervical adeno-carcinoma, and colorectal adenocarcinoma [30–34].Reports from a patient with metastatic ocular mela-noma described sustained regression of liver metasta-sis under disulfiram therapy [35]. The present studyindicates that both DSF and compound-339 inhibitcell cycle at S-phase and it has already beenestablished thatDNMT1 is transcribedpredominantlyduring the S-phase of the cell cycle where it maintainsmethylation marks after replication [36]. Apparently,the compounds act primarily during the S-phase ofcell cycle when optimal levels of DNMTs are present,arresting cell-cycle and leading to cancer cell

Figure 7. (A) Effect of compounds on cell proliferation and apoptosismarkers of ER-b deficient cells at 48 h, as analyzed by western blotting.�P< 0.05, ��P< 0.01, ��P< 0.001. (B) Proliferating cell nuclearantigen (PCNA), Bax and Bcl2 in disulfiram (DSF) and compound-339

treated ER-b siRNA transfected cells. �P< 0.05, ��P< 0.01, ��P< 0.001. (1) control; (2) scrambled siRNA; (3) ER-b siRNA; (4)compound-339 (1mM); (5) DSF (2mM); (6) compound-339 (1mM)þER-b siRNA; (7) DSF (2mM)þ ER-b siRNA.

1852 SHARMA ET AL.


apoptosis, as seen with PC3 and DU145 cells in thepresent study.DSF and compound-339 inhibited DNMT enzymes

significantly in a concentration-dependent manner,with compound-339 being almost twice more effica-cious at low concentrations thanDSF, in both PC3 andDU145. As a non-nucleoside DNMT inhibitor, DSF hasbeen found to inhibit DNMTs [20] and reduce globalgenomic methylated cytosine content by inhibitingDNMT enzyme activity. In this study, significantreduction in global DNA methylation was seen with3.0mM disulfiram and 2.0mM compound-339.ER-b is the main estrogen receptor subtype in the

prostatic tissue, which serves as a tumor suppressorgene by protecting against uncontrolled cell prolifer-ation in normal prostate [37]. Several tumor suppres-sor genes (including ER-b) have been reported to betranscriptionally silenced by hypermethylation dur-ing cancer development. Epigenetic silencing of ER-bgene by methylation has been observed in severalhuman cancers, such as breast cancer [38], lungcancer, and colorectal cancer [39]. The present study

found low level of ER-b mRNA and protein expres-sions in prostate cancer cell lines as compared withnormal prostate cell line RWPE-1 (SupplementaryFigure S1) and correlated it with aberrantmethylationof its CpG islands. Hypermethylation of CpG islandsof ER-bmay silence this gene expression, as evidencedby its re-expression by DNMT inhibitors DSF andcompound-339 in PC3 cell line in the present study.Though both the compounds significantly erasedthe methylation marks, compound-339 was far moreefficacious than DSF. Thus, in high grade prostatecancers, methylation of CpG islands within thepromoter region of the ER-b gene may lead to itstranscriptional inactivation through several unde-fined mechanisms. The concentration-dependentincrease in protein expression of ER-b and its targetgene TNF-a [40], but no changes in ER-a and its targetgene PS2, clearly demonstrate the transcriptional re-activation of ER-b gene at sub-EC50 doses by thecompounds. Since maximum expression of ER-b wasseen at 48h in both the cell lines, this time point wasused in all the subsequent experiments. ER-b is

Figure 8. (A) Viability of PC3 and DU145 cells treated with testcompounds and various combinations of ER-b agonist (DPN) and itsantagonist (THC): Disulfiram (DSF;2mM), compound-339 (1mM), DPN(100 nM or 75 nM), THC (5mM) for 24 h. (1) Control; (2) DSF; (3) 339;(4) DPN; (5) THC; (6) DSFþDPN; (7) 339þDPN; (8) DSFþ THC; (9)339þ THC; (10) DSFþDPNþ THC; (11) 339þDPNþTHC. (B) In PC3

cells, effect of test compounds and combinations of both ER-b and ER-a antagonists [THC (5mM) andMPP (7mM)] on cell viability: (a) Control;(b) THC; (c) MPP; (d) DSF (e) DSFþ THC; (f) DSFþMPP; (g) DSFþTHCþMPP; (h) 339; (i) 339þ THC; (j) 339þMPP; (k) 339þ THCþMPP. �P< 0.05, ��P< 0.01, and ��P< 0.001 versus Control, N¼ 3.



mainly located in the cytoplasm, which gets trans-located to nucleus after induction with estradiol (E2).A higher nuclear staining of ER-b in nucleus aftertreatment with compounds and E2, than with E2alone, indicates increased nuclear tanslocation of ER-

b protein in cancer cells for transactivation of targetgenes. This is further validated by the reporter geneassay, which demonstrates that the increased ER-blevels detected after compound treatments could bindeffectively to estrogen response elements (ERE) and

Figure 9. DNA methylation analysis using direct bisulfite PCRsequencing. (A) Summary of the percentage of DNA methylation at27 CpG sites of selected region for RWPE-1 cell line; and PC3 cell line inpresence and absence of test compounds. (B) Electropherogram of

rectangular area from (A) with 5 CpG sites highlighted in light greenbackground. Normal prostate cell line RWPE-1 displays approximately10% methylation at one site whereas PC3 cell line shows varyingdegrees of methylation as seen by overlapping C/T signals.

1854 SHARMA ET AL.


Figure 10. (A) Mean Relative Tumor Volume (RTV, expressed astumor volume on day of measurement/tumor volume on day 1) and (B)average animal weight of NOD-SCID mice with PC3 xenografts andtreatedwith vehicle (control), disulfiram (DSF), and compound-339. (C)

DNMT activity of tumor tissues of control and treated animals at theend of the experiment, (D) tumor tissue sections stained with eitherH&E or Ki-67 and ER-b antibodies for immunohistochemistry. Nucleiwere stained with DAPI.



transactivate the luciferase gene. Specific ER-b agonist(diarylpropionitrile, DPN) and antagonist (tetrahy-drochrysene, THC) served as reference controls.Expectedly, compound-339 was more effective thanDSF per se in transactivation of luciferase gene whilein presence of specific agonist (DPN) both DSF andcompound-339 were almost equally effective. On theother hand, inhibition of luciferase by the presence ofTHC indicated the specificity of action. It can thus beconcluded that re-expression of epigenetically re-pressed ER-b by DSF and compound-339 makes itavailable for tumor repressor action, especially inpresence of its natural agonists in vivo like 3b-Adiol [41]. Since ER-b expression is generally lost inhigh grade prostate tumors [42] and ER-b KO (and notER-aKO)mice develop prostatic cancer [43], silencingof ER-b by siRNA deregulated prostate cancer cellproliferation and apoptosis, as indicated by changesin protein expression of Bax, Bcl2, and PCNA.However, this could effectively be reversed by theaddition of compounds, plausibly due to restorationof ER-b levels.

Diarylpropionitrile (DPN) is an ER-b selectiveagonist [44] while R, R-tetrahydrochrysene (THC) isa pure ER-b antagonist with mild ER-a agonism [45],andmethyl-piperidinopyrazole (MPP) strongly antag-onises ER-a [46]. Using various combinations of DPN,THC, and MPC with DSF and compound-339 on PC3and DU145 cell lines it became evident that theactivation of ER-b leads to lower cancer cell viability.Similarly, inhibition of ER-a activity decreased cellviability, while blocking ER-b activity increased thecell number. In most of these experiments, com-pound-339 had an edge over DSF in inhibiting cancercell viability.

As an antabuse drug, DSF inhibits aldehydedehydrogenase-2 (ALDH-2) enzyme by interactingwith cysteine thiols of ALDH-2 active site. With this,assumption docking of DSF was performed onDNMT1 crystal structure, which has a CXXC motifin its auto-inhibitory linker domain. However, nostable conformation (based on docking energy) wasfound. On the other hand, the auto-inhibitorymechanism of DNMT1 involves interaction ofCXXC domain with DNA, which drives the auto-inhibitory linker to a position that prevents interac-tion between un-methylated DNA and the active siteof the MTase domain [47]. It was thus hypothesisedthat DSF may be interacting within the MTasedomain. Accordingly, we performed docking of DSFwith theMTase domain ofDNMT1. The twofold lowerdocking energy of compound-339 as compared toDSFis somewhat reflected by its �twofold higher anti-proliferative activity and DNMT inhibitory activity invitro, and�twofold higher anti-tumor activity in vivoin xenograftmicemodel, thanDSF. However, this wasnot accompanied by an increase in cyto-toxicity ofthe compound-339 over DSF. In fact in NOD- SCIDmice, the compounds did not alter the body weights

of animals during treatment, which indicated theabsence of any severe toxicity. On the other hand,decreased staining of Ki-67 was accompanied byincreased expression of ER-b in treated animal tumortissues as compared to control, and the difference inpotency of the two compounds was quite apparent.Thus the present study reaffirms the DNMT

inhibitory and anti-prostate cancer activity of DSF.It also reports the optimization of DSF molecule to adoubly potent structure (compound-339) withoutany escalation in general toxicity. This first reportcarefully examines the re-expression of epigeneticallyrepressed ER-b in prostate cancer after treatment withDSF and the new compound-339, and overtly displaysthe mechanisms that may follow to prevent cellproliferation.

ACKNOWLEDGMENTS

The authors would like to thank the SAIF Divisionof CSIR-CDRI for the NMR spectra and Flow Cytom-etry. Mr. Vikash Singh helped in docking studies. Thisstudy was supported by a grant from the Ministry ofHealth and Family Welfare, Government of India.

REFERENCES

1. Drudge-Coates L, Turner B. Prostate cancer overview. Part 1:Non-metastatic disease. Br J Nurs 2012;21:S23–S28.

2. Boyd LK, Mao X, Lu YJ. The complexity of prostate cancer:Genomic alterations and heterogeneity. Nat Rev Urol2012;9:652–664.

3. Perry AS, Watson RW, Lawler M, Hollywood D. Theepigenome as a therapeutic target in prostate cancer. NatRev Urol 2010;7:668–680.

4. Das PM, Singal R. DNA methylation and cancer. J ClinOncol2004;22:4632–4642.

5. Sharma S, Kelly TK, Jones PA. Epigenetics in cancer.Carcinogenesis 2010;31:27–36.

6. Yegnasubramanian S, Kowalski J, Gonzalgo ML, et al.Hypermethylation of CpG islands in primary and metastatichuman prostate cancer. Cancer Res 2004;64:1975–1986.

7. Yegnasubramanian S, Haffner MC, Zhang Y, et al. DNAhypomethylation arises later in prostate cancer progressionthan CpG island hypermethylation and contributes to meta-static tumor heterogeneity. Cancer Res 2008;68:8954–8967.

8. Nelson WG, De Marzo AM, Yegnasubramanian S. Epigeneticalterations in human prostate cancers. Endocrinology 2009;150:3991–4002.

9. Nelson WG, Yegnasubramanian S, Agoston AT, et al.Abnormal DNA methylation, epigenetics, and prostatecancer. Front Biosci 2007;12:4254–4266.

10. Li LC, Carroll PR, Dahiya R. Epigenetic changes in prostatecancer: Implication for diagnosis and treatment. J Natl CancerInst 2005;97:103–115.

11. Baylin SB, Esteller M, Rountree MR, Bachman KE, Schuebel K,Herman JG. Aberrant patterns of DNA methylation, chroma-tin formation, and gene expression in cancer. HumMol Genet2001;10:687–692.

12. Herman JG, Baylin SB. Gene silencing in cancer in associationwith promoter hypermethylation. N Engl J Med 2003;349:2042–2054.

13. Esteller M. Epigenetics in cancer. N Engl J Med 2008;358:1148–1159.

14. Kuiper GG, Enmark E, Pelto-Huikko M, Nilsson S, GustafssonJA. Cloning of a novel receptor expressed in rat prostate andovary. Proc Natl Acad Sci USA 1996;93:5925–5930.

1856 SHARMA ET AL.


15. Li LC, Yeh CC, Nojima D, Dahiya R. Cloning and characteriza-tion of human estrogen receptor beta promoter. BiochemBiophys Res Commun 2000;275:682–689.

16. Sasaki M, Tanaka Y, Perinchery G, et al. Methylation andinactivation of estrogen, progesterone, and androgen recep-tors in prostate cancer. J Natl Cancer Inst 2002;94: 384–390.

17. Zhu X, Leav I, Leung YK, et al. Dynamic regulation of estrogenreceptor-beta expression by DNAmethylation during prostatecancer development and metastasis. Am J Pathol 2004;164:2003–2012.

18. Nojima D, Li LC, Dharia A, et al. CpGhypermethylation ofthe promoter region inactivates the estrogen receptor-betagene in patients with prostate carcinoma. Cancer 2001;92:2076–2083.

19. Nakayama T, Watanabe M, Suzuki H, et al. Epigeneticregulation of androgen receptor gene expression in humanprostate cancers. Lab Invest 2000;80:1789–1796.

20. Lin J, Haffner MC, Zhang Y, et al. Disulfiram is a DNAdemethylating agent and inhibits prostate cancer cell growth2011. Prostate 2011;71:333–343.

21. Veverka KA, Johnson KL, Mays DC, Lipsky JJ, Naylor S.Inhibition of aldehyde dehydrogenase by disulfiram and itsmetabolite methyl diethylthiocarbamoyl-sulfoxide. BiochemPharmacol 1997;53:511–518.

22. Pradhan M, Esteve PO, Chin HG, Samaranayke M, Kim GD,Pradhan S. CXXC domain of human DNMT1 is essential forenzymatic activity. Biochemistry 2008;47:10000–10009.

23. Jung Y, Park J, Kim TY, et al. Potential advantages of DNAmethyltransferase 1 (DNMT1)-targeted inhibition for cancertherapy. J Mol Med 2007;85:1137–1148.

24. Song J, Rechkoblit O, Bestor TH, Patel DJ. Structure of DNMT1-DNA complex reveals a role for autoinhibition in maintenanceDNA methylation. Science 2011;331:1036–1040.

25. Guex N, Peitsch MC. SWISS-MODEL and the Swiss-PdbViewer: An environment for comparative protein model-ing. Electrophoresis 1997;18:2714–2723.

26. Morris GM, Huey R, Lindstrom W, et al. AutoDock4 andAutoDockTools4: Automated docking with selective receptorflexibility. J Comput Chem 2009;30:2785–2791.

27. Pettersen EF, Goddard TD, Huang CC, et al. UCSF Chimera—A visualization system for exploratory research and analysis.J Comput Chem 2004;25:1605–1612.

28. Mack GS. Epigenetic cancer therapy makes headway. J NatlCancer Inst 2006;98:1443–1444.

29. Iljin K, Ketola K, Vainio P, et al. High-throughput cell-basedscreening of 4910 known drugs and drug-like small moleculesidentifies disulfiram as an inhibitor of prostate cancer cellgrowth. Clin Cancer Res 2009;15:6070–6078.

30. Wickstrom M, Danielsson K, Rickardson L, et al. Pharmaco-logical profiling of disulfiramusing human tumor cell lines andhuman tumor cells from patients. Biochem Pharmacol2007;73:25–33.

31. Chen D, Cui QC, Yang H, Dou QP. Disulfiram, a clinically usedanti-alcoholism drug and copper-binding agent, inducesapoptotic cell death in breast cancer cultures and xenograftsvia inhibition of the proteasome activity. Cancer Res2006;66:10425–10433.

32. Lovborg H, Oberg F, Rickardson L, Gullbo J, Nygren P, LarssonR. Inhibition of proteasome activity, nuclear factor-KappaBtranslocation, and cell survival by the anti-alcoholism drugdisulfiram. Int J Cancer 2006;118:1577–1580.

33. Cho HJ, Lee TS, Park JB, et al. Disulfiram suppresses invasiveability of osteosarcoma cells via the inhibition of MMP-2and MMP-9 expression. J Biochem Mol Biol 2007;40:1069–1076.

34. Wang W, McLeod HL, Cassidy J. Disulfiram-mediatedinhibition of NF-kappaB activity enhances cytotoxicity of 5-fluorouracil in human colorectal cancer cell lines. Int J Cancer2003;104:504–511.

35. Brar SS, Grigg C,Wilson KS, et al. Disulfiram inhibits activatingtranscription factor/cyclic AMP-responsive element bindingprotein and human melanoma growth in a metal-dependentmanner in vitro, in mice and in a patient with metastaticdisease. Mol Cancer Ther 2004;3:1049–1060.

36. Robertson KD, Keyomarsi K, Gonzales FA, Velicescu M, JonesPA. Differential mRNA expression of the human DNAmethyltransferases (DNMTs) 1, 3a, and 3b during the G(0)/G(1) to S phase transition in normal and tumor cells. NucleicAcids Res 2000;28:2108–2113.

37. Zhang X, Leung YK, Ho SM. AP-2 regulates the transcriptionof estrogen receptor (ER)-beta by acting through a methyl-ation hotspot of the 0N promoter in prostate cancer cells.Oncogene 2007;26:7346–7354.

38. Ottaviano YL, Issa JP, Parl FF, Smith HS, Baylin SB, DavidsonNE. Methylation of the estrogen receptor gene CpGislandmarks loss of estrogen receptor expression in human breastcancer cells. Cancer Res 1994;54:2552–2555.

39. Issa JP, Ottaviano YL, Celano P, Hamilton SR, Davidson NE,Baylin SB. Methylation of the oestrogen receptor CpG islandlinks ageing and neoplasia in human colon. Nat Genet1994;7:536–540.

40. McPherson SJ, Hussain S, Balanathan P, et al. Estrogenreceptor-beta activated apoptosis in benign hyperplasiaand cancer of the prostate is androgen independent andTNFalpha mediated. Proc Natl Acad Sci USA 2010;107:3123–3128.

41. Muthusamy S, Andersson S, Kim HJ, et al. Estrogen receptorbeta and 17beta-hydroxysteroid dehydrogenase type 6, agrowth regulatory pathway that is lost in prostate cancer. ProcNatl Acad Sci USA 2011;108:20090–20094.

42. Bardin A, Boulle N, Lazennec G, Vignon F, Pujol P. Loss ofERbeta expression as a common step in estrogen-dependenttumor progression. Endocr Relat Cancer 2004;11:537–551.

43. Ricke WA, McPherson SJ, Bianco JJ, Cunha GR, Wang Y,Risbridger GP. Prostatic hormonal carcinogenesis is mediatedby in situ estrogen production and estrogen receptor alphasignaling. Faseb J 2008;22:1512–1520.

44. Meyers MJ, Sun J, Carlson KE, Marriner GA, Katzenellenb-ogen BS, Katzenellenbogen JA. Estrogen receptor-betapotency-selective ligands: Structure-activity relationship stud-ies of diarylpropionitriles and their acetylene and polaranalogues. J Med Chem 2001;44:4230–4251.

45. Meyers MJ, Sun J, Carlson KE, Katzenellenbogen BS,Katzenellenbogen JA. Estrogen receptor subtype-selectiveligands: Asymmetric synthesis and biological evaluation of cis-and trans-5, 11-dialkyl- 5,6,11, 12-tetrahydrochrysenes.J Med Chem 1999;42:2456–2468.

46. Sun J, Huang YR, HarringtonWR, Sheng S, KatzenellenbogenJA, Katzenellenbogen BS. Antagonists selective for estrogenreceptor alpha. Endocrinology 2002;143:941–947.

47. Godley LA, Mondragon A. Molecular biology. Preference byexclusion. Science 2011;331:1017–1018.

SUPPORTING INFORMATION

Additional supporting information may be found inthe online version of this article at the publisher’sweb-site.



本文献由“学霸图书馆-文献云下载”收集自网络，仅供学习交流使用。

学霸图书馆（www.xuebalib.com）是一个“整合众多图书馆数据库资源，

提供一站式文献检索和下载服务”的24 小时在线不限IP

图书馆。

图书馆致力于便利、促进学习与科研，提供最强文献下载服务。

图书馆导航：

图书馆首页文献云下载图书馆入口外文数据库大全疑难文献辅助工具

http://www.xuebalib.com/cloud/

http://www.xuebalib.com/

http://www.xuebalib.com/cloud/


http://www.xuebalib.com/vip.html

http://www.xuebalib.com/db.php

http://www.xuebalib.com/zixun/2014-08-15/44.html


disulfiram and its novel derivative sensitize prostate ...download.xuebalib.com/37a8wskz3cxh.pdf ·...

Documents