P

Lm

BQYZHa

b

c

d

a

ARRAA

C2P

KLEGML

scnmcCIdN

v

(

a

h1

Pharmacological Research 122 (2017) 66–77

Contents lists available at ScienceDirect

Pharmacological Research

j ourna l h o mepa ge: www.elsev ier .com/ l ocate /yphrs

erspective

PE-1, an orally active pyrimidine derivative, inhibits growth andobility of human esophageal cancers by targeting LSD1

o Wang a,b,c,d,1, Bing Zhao a,b,c,d,1, Lu-Ping Pang a,b,c,d,1, Yuan-Di Zhao a,b,c,d,ian Guo a,b,c,d, Jun-Wei Wang a,b,c,d, Yi-Chao Zheng a,b,c,d, Xin-Hui Zhang a,b,c,d,ing Liu a,b,c,d, Guang-Yao Liu a,b,c,d, Wen-Ge Guo a,b,c,d, Chao Wang a,b,c,d,hong-Hua Li a,b,c,d, Xue-Jing Mao a,b,c,d, Bin Yu a,b,c,d,∗, Li-Ying Ma a,b,c,d,∗,ong-Min Liu a,b,c,d,∗

School of Pharmaceutical Sciences, Zhengzhou University, Zhengzhou, ChinaCollaborative Innovation Center of New Drug Research and Safety Evaluation, Henan Province, ChinaKey Laboratory of Technology of Drug Preparation (Zhengzhou University), Ministry of Education, ChinaKey Laboratory of Henan Province for Drug Quality and Evaluation, China

r t i c l e i n f o

rticle history:eceived 21 February 2017eceived in revised form 19 May 2017ccepted 25 May 2017vailable online 30 May 2017

hemical compound:-Phenylcyclopropanamine (2-PCPA)ubChem CID: 5530

a b s t r a c t

Histone lysine specific demethylase 1 (LSD1) plays an important role in epigenetic modifications, andaberrant expression of LSD1 predicts tumor progression and poor prognosis in human esophageal cancers.In this study, a series of LSD1 inhibitors were synthesized and proved to be highly potent against humanesophageal squamous cell carcinoma (ESCC). Our data showed that these LSD1 inhibitors selectivelysuppressed the viability of esophageal cancer cell line (EC-109) bearing overexpressed LSD1. Amongthese, compound LPE-1 (LSD1 IC50 = 0.336 ± 0.003 �M) significantly suppressed proliferation, inducedapoptosis, arrested cell cycle of EC109 cells at G2/M phase, and caused changes of the associated proteinmarkers correspondingly. We also found that compound LPE-1 potently inhibited the migration andinvasion of EC-109 cells. Docking studies showed that the cyano group formed hydrogen bonds with

eywords:PE-1sophageal cancersrowthobility

SD1

Val811 and Thr810. Additionally, the thiophene moiety formed arene-H interaction with Trp761 residue.In vivo studies showed that compound LPE-1 inhibited tumor growth of xenograft models bearing EC-109 without obvious toxicity. Collectively, our findings indicate that LSD1 may be a potential therapeutictarget in ESCC, and compound LPE-1 could serve as a lead compound for further development for anti-ESCC drug discovery.

© 2017 Elsevier Ltd. All rights reserved.

Abbreviations: LSD1, histone lysine specific demethylase 1; ESCC, esophagealquamous cell carcinoma; E2F1, E2F transcription factor; FAD, flavin adenine dinu-leotide; ERBB2, epidermal growth factor receptor 2; CCNA2, cyclin-A2; NSCLC,on-small cell lung cancer; PHLDB2, Pleckstrin homology-like domain family Bember 2; EMT, epithelial-mesenchymal transition; HPLC, high performance liquid

hromatography; RPMI, Roswell Park Memorial Institute; FBS, fetal bovine serum;CTCC, the China Center for Type Culture Collection; DMSO, dimethyl sulfoxide;

C50, 50% inhibitory concentration; SDS, sodium dodecyl sulfate; SDS-PAGE, sodiumodecyl sulfate polycrylamide gel electrophoresis; ZEB, zine-finger E-box-binding;TA, nitrilotriacetic acid.∗ Corresponding authors at: School of Pharmaceutical Sciences, Zhengzhou Uni-ersity, Zhengzhou, China

E-mail addresses: zzuyubin@hotmail.com (B. Yu), maliying@zzu.edu.cnL.-Y. Ma), liuhm@zzu.edu.cn (H.-M. Liu).

1 These authors contributed equally to the work and should be regarded as equiv-lent first authors.

ttp://dx.doi.org/10.1016/j.phrs.2017.05.025043-6618/© 2017 Elsevier Ltd. All rights reserved.

1. Introduction

Epigenetics is typically defined as the study of heritable changesin gene expression that are not due to changes in DNA sequence. Inthe past two decades, the epigenetic post-transcriptional modifi-cation mostly focused on the DNA, RNA, and histone modifications[1]. Among them, histone modifications, including methylation,acetylation, phosphorylation, ubiqutination, sumoylation and soon, associate with the gene transcription or not [2]. Historically,histone methylation was believed to be an irreversible process untilthe histone lysine specific demethylase 1 (LSD1, also known asKDM1A) was first identified by Shi Yang’s group in 2004 [3]. LSD1 isan flavin adenine dinucleotide (FAD) dependent oxidative enzyme

and specifically demethylates the mono- and di-methylated K4 andK9 of histone 3 [3,4], as well as some non-histone proteins such as

gical R

E(

ehssg[tnltpis(mmu[csc

itaLpGtiixtcd

2

2

2

sd1

srTtr

22tAwep

2od(

B. Wang et al. / Pharmacolo

2F transcription factor (E2F1) [5], p53 [6], DNA methyltransferasesDNMTs) [4,7].

LSD1 has been proved to be tightly associated with tumorigen-sis. Lim [8] et al. found that LSD1 was obviously overexpressed inuman estrogen-receptor-negative breast cancer tissues and LSD1iRNA-induced LSD1 knockdown could increase the expression ofome proliferation-related genes, including p21, human epidermalrowth factor receptor 2 (ERBB2), and cyclin-A2 (CCNA2). Lv et al.9] reported that LSD1 was overexpressed in lung cancer tissueshan normal lung tissues, and the level of LSD1 expression wasegatively associated with the total survival time of non-small cell

ung cancer (NSCLC) patients. Yu [10] and Chen [11] demonstratedhat high expression of LSD1 was correlated with poor prognosis ofatients with ESCC. Isamu Hoshino [12] further indicated that LSD1

nhibitors can inhibit cell growth by regulating some gene expres-ion, such as Pleckstrin homology-like domain family B member 2PHLDB2), in ESCC. Our group reported two novel series of small-

olecular LSD1 inhibitors that markedly suppressed invasion andigration of LSD1 overexpressed gastric cancer cells through reg-

lating the process of epithelial-mesenchymal transition (EMT)13,14]. LSD1 overexpression has also been observed in prostaticancer [15], colon cancer [16], bladder cancer [17] et al. These datauggest that LSD1 may be a promising therapeutic drug target forancer therapy.

Following our previous success in identifying new LSD1nhibitors for anticancer treatment [13,14,18–23], we herein reporthe synthesis and antitumor activity of a series of LSD1 inhibitorsgainst ESCC as well as the potential mechanisms. The most potentSD1 inhibitor, compound LPE-1, selectively inhibited LSD1, sup-ressed proliferation, induced apoptosis, arrested the cell cycle at2/M phase of EC109, and caused changes of the associated pro-

ein markers. Compound LPE-1 also suppressed the migration andnvasion of EC-109 through inhibiting EMT process. In vivo stud-es showed that compound LPE-1 inhibited the growth of EC-109enograft tumors without obvious toxicity. Our findings indicatehat LSD1 may be a potential therapeutic target in ESCC, and theompound LPE-1 could be a lead compound for further anti-ESCCrug discovery.

. Materials and methods

.1. Chemical

.1.1. GeneralReagents and solvents were purchased from commercial

ources and used without further purification. Melting points wereetermined on an X-5 micromelting apparatus, and 1H NMR and3C NMR spectrum were recorded on a Bruker 400 and 100 MHzpectrometer, respectively. High-resolution mass spectra wereecorded on a Waters Micromass Q-T of Micromass spectrometer.he purity of all biologically evaluated compounds was determinedo be >95% by reverse phase high performance liquid chromatog-aphy (HPLC) analysis.

.1.2. Synthesis and characterization

.1.2.1. General procedure for the synthesis of compounds 4a-l. Thearget compounds were synthesized as previously described [14].s shown in Scheme 1, the 6-thiazole-5-cyano-2-thiouracil 4a-lere prepared from aromatic aldehydes 1 (1 mmol), ethylcyanoac-

tate 2 (1 mmol), and thiourea 3 (1 mmol) in the presence ofotassium carbonate in ethanol.

.1.2.2. General procedure for the synthesis of compounds 5a-l. With-ut further purification, a mixture of the appropriate 2-mercapto-ihydroyrimidine derivatives 4a-l (1 mmol), the propargyl bromide1 mmol), and anhydrous potassium carbonate (1 mmol) was

esearch 122 (2017) 66–77 67

refluxed in dry dioxane. Upon completion, as judged by TLC,phosphorous oxychloride was added dropwise with stirring whilemaintaining the temperature of the reaction mixture. Stirring wascontinued for additional 1 h. The cooled reaction mixture waspoured on crushed ice and the separated solid was filtered off,washed with water, dried and crystallized from aqueous ethanolto yield the pure product.

2.1.2.3. General procedure for the synthesis of compounds 6a-l. To awell stirred solution of the thiosemicarbazide (1 mmol) in abso-lute ethanol (5 mL), equimolar amount of a solution of compounds5a-l (1 mmol) in absolute ethanol (2 mL) was added. The reactionmixture was stirred and refluxed for 5 h. Upon completion, theprecipitated product was filtered off and washed with ethanol toafford the crude product. The crude product was recrystallized fromethanol to yield the pure product.

Compound 6l is also named as LPE-1.

2.1.2.4. 4-chloro-2-(prop-2-yn-1-ylthio)-6-(thiophen-2-yl)pyrimidine-5-carbonitrile (compound 5l). Yield 52%. Yellowsolid. Mp: 119–120 ◦C. 1H NMR (400 MHz, CDCl3, �, ppm) ı 8.52(dd, J = 4.0, 0.9 Hz, 1H, Ar-H), 7.78 (dd, J = 5.0, 0.9 Hz, 1H, Ar-H), 7.27(dd, J = 4.9, 4.1 Hz, 1H, Ar-H), 3.99 (d, J = 2.6 Hz, 2H, CH2 ), 2.26 (t,J = 2.6 Hz, 1H, C H). 13C NMR (100 MHz, CDCl3, �, ppm): ı 8.45,8.42, 8.30, 8.27, 4.01, 4.00, 2.30, 2.29, 2.29. HR-MS (ESI): Calcd.C12H7ClN3S2, [M + H]+m/z: 291.9770; found: 291.9771.

2.1.2.5. 2-(5-Cyano-2-(prop-2-yn-1-ylthio)-6-(2-Thienyl)-pyrimidin-4-yl)hydrazine-carbothioamide (compound 6l, LPE-1).Yield 82%. Yellow solid. Mp: 204–205 ◦C. 1H NMR (400 MHz,DMSO-d6, �, ppm) ı 10.20 (s, 1H, pyrimidine−NH, D2O exchange-able), 9.53 (s, 1H, thiourea−NH, D2O exchangeable), 8.26 (d,J = 2.7 Hz, 1H, Ar-H), 8.01 (d, J = 4.4 Hz, 1H, Ar-H), 7.95 (s, 1H,thiourea−NH2, D2O exchangeable), 7.76 (s, 1H, thiourea−NH2,D2O exchangeable), 7.42–7.24 (m, H, Ar−H), 4.02 (d, J = 2.5 Hz,2H, CH2 ), 3.17 (t, J = 2.6 Hz, 1H, C H). 13C NMR (100 MHz,DMSO-d6, �, ppm) ı182.64, 172.00, 163.11, 158.83, 140.12, 134.20,131.39, 129.62, 115.93, 80.13, 74.13, 19.61. HR-MS (ESI): calcd.C13H11N6S3, [M + H]+ m/z: 347.0207; found: 347.0210.

2.2. Biological studies

2.2.1. Cell cultureHuman esophageal cancer cell lines and human esophageal

normal line were maintained in Roswell Park Memorial Institute(RPMI) 1640 medium (Hyclone, USA) with 10% fetal bovine serum(FBS) (Hyclone, USA) and 1% penicillin-streptomycin in a humid-ified atmosphere of 5% CO2 and 95% air at 37 ◦C. All cell lineswere purchased from the China Center for Type Culture Collection(CCTCC, China). For future pharmacological investigations, the com-pounds were prepared to the stock concentration of 10 mg/ml withdimethyl sulfoxide (DMSO), and the highest DMSO concentrationin the medium (0.1%) did not have any substantial effect on thedetermined cellular functions.

2.2.2. Cell viability assayExponentially growing determined cells were seeded into 96-

wells plate (Nest Biotechnology, China) at 4 × 103 cells per well.After 24 h incubation, remove the culture medium and replace freshmedium containing the candidate compound at gradient concen-trations. The cells were incubated for another 72 h. Then, 20 �L ofMTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bro-

mide) solution (5 mg/ml) was added to all wells and the cells wereincubated for 4 h at 37 ◦C. Discard the medium containing MTT,replace 150 �L DMSO, and agitate the plate until the dark violet(formazan) completely dissolved. The absorbance was measured

68 B. Wang et al. / Pharmacological Research 122 (2017) 66–77

S 10 h;

c

uttdt

2

wActCAc

2

awLwBtf

2

siwTatolqtm

2

Lt4sapa

2

Bt

cheme 1. Reagents and conditions: a: absolute ethanol, anhydrous K2CO3, reflux,: appropriate aniline, absolute ethanol, reflux, 6 h.

sing a microplate reader at a wavelength of 570 nm. Each concen-ration was analyzed in triplicate and the experiment was repeatedhree times. The average 50% inhibitory concentration (IC50) wasetermined from the concentration response curves according tohe inhibition ratio for each concentration.

.2.3. Colony formation assayOnly about 1000 exponentially growing determined cells per

ell were seeded into 6-wells plate (Nest Biotechnology, China).fter 24 h, the cells were incubated with different concentrations ofompound LPE-1, for another 14days. Then, the cells were washedwice with PBS and fixed with 4% paraformaldehyde for 30 min.rystal violet staining was used to stain the cell for 30 min at 37 ◦C.fter that, wash away the staining with PBS until the colonies werelear enough.

.2.4. Flow cytometric analysis of apoptotic analysisCells were seeded into 6-wells plate (Nest Biotechnology, China)

t 2 × 105 cells per well. After 24 h incubation, the cultured mediumas removed and replaced with fresh medium containing the

PE-1 at different concentrations for another 48 h. Then, the cellsere harvested and the AnnexinV-FITC/PI apoptosis kit (KeyGEN

ioTECH, USA) was used according to the manufacturer’s instruc-ions to detect apoptosis cells. Ten thousand events were collectedor each sample and analyzed by (BD Bioscience, USA).

.2.5. Hoechst 33258 staining2 × 105 per well exponentially growing determined cells were

eeded in a 6-wells plate (Nest Biotechnology, China). After 24 hncubation, the cultured medium was replaced of the fresh medium

ith different concentrations of compound LPE-1 for another 48 h.hen, remove the medium, wash the cells with cold PBS gently,nd fix the cells with 4% paraformaldehyde for 30 min. Cells werehen stained with 5ug/mL Hoechst-33258 (Beyotime Biotechnol-gy, China) in PBS containing 0.5% Triton X-100 in the dark for at

east 30 min. After remove the dye liquor and wash the cells ade-uately, apoptosis cells were examined and identified accordingo the condensation and fragmentation of nuclei by fluorescence

icroscopy under UV excitation.

.2.6. Flow cytometric analysis of cell cycle distributionAfter incubation with different concentrations of compound

PE-1 for 48 h, cells were harvested and washed with cold PBS forwo times. Then, the cells were fixed with 70% ice-cold ethanol at◦C overnight. The fixed cells were washed twice with cold PBS and

tained with 1 mL PBS solution containing 1% Triton X-100, RNasend PI (KeyGEN BioTECH, USA) for 30 min under dark condition asrevious. Ten thousand events were collected for each sample andnalyzed by (BD Bioscience, USA).

.2.7. siRNA transfectionAbout 1 × 107 EC-109 cells were seeded in a 6-well plate (Nest

iotechnology, China) and then incubated for another 48 h untilhe cells were 60–80% confluent at the time of transfection. Then

b: (i) propargyl bromide, dioxane, reflux; (ii) phosphorous oxychloride, reflux, 1 h;

prepare RNA-lipid complexes and add them to cells according tothe manufacture’s instructions (GenePharma, China).

2.2.8. Migration and invasion assayWound healing assay, transwell assay and matrigel-coated tran-

swell assay were conducted to evaluate the cell migration andinvasion ability. All these experiments were carried out as previ-ously published [13,14].

For the wound healing assay, cells were seeded in a 24-wellsplate (CORNING, USA) at 1 × 104 per well. After 24 h incubation, thecell surface was scratched using a 10 �L pipet tip. Then, the cellswere cultured with fresh medium containing 1% FBS and differentconcentrations of compound LPE-1 for 48 h, and photographed onan inverted microscope.

For migration assay, cells were seeded in Transwell 24-wellsplate (CORNING, USA). 1% heat-inactivated FBS and 20% FBS wereadded into upper and lower chamber separately. Different con-centrations of compound LPE-1 were added in the chamber. After48 h incubation, remove the medium, wash the chambers withPBS and the migrating cells were fixed with methanol for 15 minand stained with Hoechst-33258. Each chamber was photographedusing Thermo Fisher Cellomics High Content System.

For the matrigel-coated transwell assay, cells were seeded inTranswell 24-wells plate with Matrigel (BD Bioscience, USA). Themedium, attractant, staining and cell counting method were thesame as those of the Transwell assay.

2.2.9. Western blotting analysisCells were cultured with different concentrations of compound

LPE-1 for 48 h, and then the floating cells were collected andwashed with cold PBS. Western blot was performed with the totallysates by cell lysis buffer [1% NP-40, 0.1% sodium dodecyl sulfate(SDS), 150 mM NaCl, 25 mM Tris-HCl, 1% deoxycholic acid sodiumsalt, 1% PMSF] or histone purified with a kit (Epigentek, USA).Equal amounts of cell lysates were denatured, separated by sodiumdodecyl sulfate polycrylamide gel electrophoresis (SDS-PAGE), andtransferred to 0.22 �m nitrocellulose membrane. The membranewas blocked with PBS containing 5% nonfat milk for 2 h at roomtemperature and then incubated with primary antibody overnightat 4 ◦C, followed by incubation with a horseradish peroxidase-conjugated secondary antibody. The immunoblots were visualizedby enhanced chemiluminescence kit (Thermo Fisher, USA).

Antibodies used were against histone H3K4me (ab194678,abcam, USA), H3K4me2 (ab194678, abcam, USA), H3K9me2(ab194680, abcam, USA), H3K4me3 (ab71998, abcam, USA), totalH3 (E1A6359, EnoGene Biotechnology, China), total H4 (E1A6355,EnoGene Biotechnology, China), pro-caspase3 (#9665, Cell Sig-naling, USA), cleaved-caspase3 (#9664, Cell Signaling, USA),pro-caspase7 (#12827, Cell Signaling, USA), cleaved-caspase7(#8438, Cell Signaling, USA), Bcl-2 (E1A6139, EnoGene Biotechnol-

ogy, China), Bax (E1A0120, EnoGene Biotechnology, China), LSD1(ab37165, abcam, USA), Snai1 (#3879, Cell Signaling, USA), Slug(#9585, Cell Signaling, USA), N-Cadherin (#13116, Cell Signaling,USA), Vimentin (#5741, Cell Signaling, USA), E-Cadherin (#3195,

B. Wang et al. / Pharmacological Research 122 (2017) 66–77 69

Fig. 1. The LSD1expression levels of four human cell lines: human normal esophageal cell line (Het-1A) and human esophageal cancer cell lines (EC-109, TE-1, and EC-9706).

Table 1Inhibitory results of preliminary evaluation against LSD1 in vitro and four human cell lines for the target compounds.

Compd. Ar X LSD1(�M)a IC50(�g/mL)

Het-1A EC-109 TE-1 EC-9706

6a S 1.59 ± 0.22 >32 13.85 ± 0.55 20.76 ± 0.61 >32

6b S 0.65 ± 0.12 >32 6.76 ± 0.52 9.94 ± 0.69 >32

6c S 1.10 ± 0.93 >32 13.99 ± 0.98 16.60 ± 1.25 26.56 ± 2.11

6d S 1.08 ± 0.03 >32 5.66 ± 0.45 7.30 ± 0.49 >32

6e S 1.82 ± 0.26 >32 12.70 ± 0.89 19.84 ± 1.65 22.87 ± 1.96

6f S 1.59 ± 0.20 >32 26.74 ± 0.70 30.08 ± 0.71 >32

6g S 1.66 ± 0.22 >32 14.37 ± 0.60 22.67 ± 0.67 33.84 ± 0.74

6h S 1.59 ± 0.20 >32 14.49 ± 0.64 24.66 ± 0.74 >32

6i O 2.64 ± 0.42 >32 8.65 ± 0.50 26.12 ± 0.67 >32

6j O 4.88 ± 0.68 >32 12.64 ± 0.56 14.45 ± 0.59 19.09 ± 0.62

6k O 5.06 ± 0.70 >32 23.60 ± 0.65 25.04 ± 0.66 >32

6l(LPE-1) S 0.34 ± 0.01 >32 5.76 ± 0.43 22.76 ± 0.63 >322-PCPA – – 27.83 ± 2.64 >32 >32 >32 >32

H at ina ofiles.

CG

2

begU5

et-1A, EC-109, TE-1 and EC-9706 cells were treated with compounds 6a-l for 72 hssays were used to monitor the IC50 values for determining the drug sensitivity pr

a Data presented are adapted from our previous paper [14].

ell Signaling, USA, USA), Occludin (#5506, Cell Signaling, USA),APDH (E1A7021, EnoGene Biotechnology, China).

.2.10. Xenograft studyAnimals were treated according to the protocols established

y the ethics committee of Zhengzhou University and the in vivo

xperiments were carried out in accordance with the approveduidelines and approved by the ethics committee of Zhengzhouniversity. Female BALB/c nude mice (weighing 18–20 g and aged–6 weeks at the beginning of the assay) were purchased from

creasing concentrations (0, 1, 2, 4, 8, 20, 40, 80, 100 �g/ml). The MTT colorimetric

Hunan Slack Scene of Laboratory Animal Co., Ltd. (Hunan, China).Xenograft model using human esophageal cancer cell lines, EC109,were established in BALB/c mice. Until the volume of tumorsreached 100 mm3, mice were separated into control group andtreatment group. The treatment group received compound LPE-1 at 50 mg/kg, p.o. per day for a period of 21 days during which the

body weight was measured and tumor volume was determinedby vernier caliper measurement at 3-day intervals. After the lastday, the mice were euthanized and the tumors were isolated andweighted.

70 B. Wang et al. / Pharmacological Research 122 (2017) 66–77

Fig. 2. Effects of compound LPE-1 on the proliferation of Het-1A and EC-109. (A) Het-1A and EC-109 cells were treated with compound LPE-1 at different concentrationsfor 24 h, 48 h, 72 h, respectively, then the MTT was used to evaluate the its effect on cell viability; (B) The colony formation of Het-1A and EC-109 cells after treatment ofcompound LPE-1 at different concentrations for 7 days; (C) Effect of compound LPE-1 on the cell cycle distribution of EC-109. The experiments were performed three times,and a representative experiment is shown.

F reatmo *P < 0m

2

cHgos

2

w

ig 3. The expression level of LSD1 and histone methylation in EC-109 cells after tf histone 3 (H3) and histone 4 (H4) were used as loading control. * P < 0.05 and *ean ± SD. All experiments were carried out at least three times.

.2.11. Molecular docking studyThe protein complex was obtained from protein data bank (PDB

ode: 5L3E), and all water and ligand molecules were deleted.ydrogen and partial charges were added by the protonate 3D pro-ram of MOE2014; Energy minimization of the ligands was carriedut using energy minimize program of MOE. Default parameterettings generated by the program of MOE were used for docking.

.2.12. Inhibitory evaluation of the compounds against LSD1Inhibitory effects of the candidate compounds against LSD1

ere evaluated as previously published [13]. Briefly, cDNA encod-

ent with compound LPE-1 were determined by Western Blot, and the total levels.01 were considered statistically significant compared with the control. Dates are

ing LSD1(157–852AA) was cloned into pET28b to construct theplasmid pET28b-LSD1, and then the plasmid was transfected intoBL21(DE). The recombinant was induced with 0.25 mM isopropyl�-d-1-thiogalactopyranoside (IPTG) at 20 ◦C and purified with Ni-NTA resin. The compounds were incubated with the recombinantand H3K4me2, and the fluorescence was measured at an excitationwavelength of 530 nm and an emission wavelength of 590 nm with

the addition of Amplex Red and horseradish peroxidase in order toevaluate the inhibition rate of the candidate compounds.

B. Wang et al. / Pharmacological Research 122 (2017) 66–77 71

F L3E). Pb

3

3

lfaBeleeocci(aimec(1cEl

awtocEflGm

ig. 4. Binding models of compound LPE-1 in the active site of LSD1 (PDB code: 5inding residues are highlighted in different colors.

. Results

.1. Antiproliferative activity

Compounds LPE-1 and 6a-k (compounds from our in-houseibrary [14]) were evaluated for their cytotoxic activity againstour human cell lines: human normal esophageal cell line (Het-1A)nd human esophageal cancer cell lines (EC-109, TE-1, EC-9706).efore testing the cytotoxic effects of these compounds, we firstxamined the expression levels of LSD1 in four different cellines. As shown in Fig. 1, EC-109 cells possessed the highest LSD1xpression levels, followed by TE-1, EC-9706, Het-1A. We nextvaluated the LSD1 inhibitory effect and antiproliferative activityf compounds 6a-k and LPE-1 against EC-109, TE-1 and EC-9706ells and normal esophageal cell line Het-1A using 2-PCPA as theontrol drugs. As represented in Table 1, all compounds inhib-ted LSD1 at low micromolar levels, more potent than 2-PCPAIC50 = 27.83 �M). Compound LPE-1 had improved LSD1 inhibitoryctivity (IC50 = 0.336 �M) compared to compounds 6a-k, suggest-ng the replacement of phenyl ring with heterocyclic ring system

ay be beneficial for the activity. Interestingly, the antiprolif-rative activity of these compounds against the tested cancerells correlated with the LSD1 inhibitory effect. Compound LPE-1IC50 = 0.336 �M) displayed the best inhibitory activity against EC-09 cells, which had the highest LSD1 expression levels. In contrast,ompound LPE-1 showed relatively weak inhibition toward TE-1,C-9706 and Het-1A cells bearing relatively low LSD1 expression

evels.As shown in Fig. 2A, compound LPE-1 inhibited the prolifer-

tion of EC-109 in concentration- and time-dependent manner,hile Het-1A was less sensitive against LPE-1. The difference in

he antiproliferative activity promoted us to investigate the effectf LPE-1 on the colony formation. The results EC-109 showed thatompound LPE-1 significantly inhibited proliferation of the single

C-109 cell, but was less sensitive toward Het-1A (Fig. 2B). Furtherow cytometric analysis showed that compound LPE-1 induced2/M phase arrest of EC-109 cell line in a concentration-dependentanner performed by (Fig. 2C).

utative 2D (A) and 3D (B) binding models of LPE-1 in the active site of LSD1. Key

3.2. Inhibition of compound LPE-1 on LSD1 activity

The favorable potency of compound LPE-1 against LSD1(IC50 = 0.336 �M) encouraged us to investigate the effect of com-pound LPE-1 on LSD1 substrates in EC-109 (Fig. 3). After treatmentfor 48 h, the methylation levels of LSD1 substrates, H3K4me1/me2and H3K9me2, increased dose-dependently, while H3K4me3,which is not the substrate of LSD1, was almost unchanged. What’smore, the expression of LSD1 was also not affected. Those resultssuggest that the compound LPE-1 could inhibit the LSD1 activity invitro, but not affect its expression.

3.3. Molecular docking studies

The favorable potency of compound LPE-1 toward LSD1 pro-moted us to perform the in silicon docking studies to rationalizethe potency using the MOE2014 software (PDB code: 5L3E) [24].The 2D interaction diagram (Fig. 4A) indicated that the cyano groupformed hydrogen bonds with Val811 and Thr810. Additionally, thethiophene moiety formed arene-H interaction with Trp761 residue.The 3D binding models of compound LPE-1 in the active sited ofLSD1 was represented in Fig. 4B. The docking results may providea basis for further optimization.

3.4. Effect of compound LPE-1 on apoptosis and involvedmechanism

To explore the apoptotic effect of compound LPE-1 on EC-109and TE-1, we first examined the cell morphology changes after 48 hincubation with compound LPE-1 at indicated concentrations usingHoechst 33258 staining. The typical apoptotic markers, includ-ing cell rounding, chromatin shrinkage and apoptotic bodies wereobserved, especially at high concentrations (Fig. 5A). Next, we alsoperformed apoptotic analysis with Annexin V-FITC/PI double stain-

ing and quantitated by flow cytometry. As shown in Fig. 5B, thepercentage of apoptotic EC-109 cell (including early phase and latephase apoptosis) were increased up to 22.3%, 64.9% and 68.6% atthe indicated concentrations, respectively, compared to the con-

7 gical R

tttscb

FcpLc

2 B. Wang et al. / Pharmacolo

rol (0.4%). While the Het-1A cell didn’t show obvious apoptosis athe same concentrations, and the TE-1 cells, expressing LSD1 lesshan EC-109, needed higher concentrations of LPE-1 to obtain the

ignificant apoptosis rate. Taken together, these date support theonclusion that the LPE-1 induced the apoptosis of EC-109 and TE-1y targeting LSD1.

ig. 5. Compound LPE-1 induced apoptosis in EC-109 and TE-1 cells. (A) Apoptosis analyells; (B) Quantitative analysis of apoptotic cells using Annexin V-FITC/PI double staininro- and cleaved-caspase7, Bcl-2 and Bax were determined after 48 h treatment with comSD1 knockdown with siRNA for 48 h * P < 0.05 and **P < 0.01 were considered statisticallyarried out at least three times.

esearch 122 (2017) 66–77

To further explore the mechanism of compound LPE-1 induc-ing apoptosis, the expression of key proteins involved themitochondria-related apoptosis pathway was examined by West-

ern Blot (Fig. 5C). The pro-apoptosis protein, Bax, was up-regulatedin a concentration-dependent manner, while the anti-apoptosisprotein Bcl-2 was down-regulated. EC-109 cells with down-

sis with Hoechst-33258 staining after 48 h of compound LPE-1 in EC-109 and TE-1g and flow-cytometry calculation; (C) Expressions of pro- and cleaved-caspase3,pound LPE-1. (D) Expression levels of LSD1, Bcl-2, and Bax were determined after

significant compared with the control. Dates are mean ± SD. All experiments were

B. Wang et al. / Pharmacological Research 122 (2017) 66–77 73

(Cont

ru3Lm

3

mtttcaopt

Fig. 5.

egulated LSD1 expression levels caused by siRNA also showedpregulated Bax and down-regulated Bcl-2. Meanwhile, caspase-/-7 activation were also observed after treatment with compoundPE-1. These findings indicted that LSD1 may be involved in theitochondria-related apoptosis.

.5. Effect of compound LPE-1 on cell migration and invasion

As reported previously, LSD1 was involved in epithelial-esenchymal transition (EMT), which contributed for the migra-

ion and invasion abilities of cancer cells [25,26]. So we evaluatedhe migration ability of EC-109 and TE-1 by wound healing andranswell assay, the invasion ability was evaluated by matrigel-oated transwell. As shown in Fig. 6A, after scratched the EC-109

nd TE-1 cells face, compound LPE-1 inhibited the would healingbviously, the transwell assay (Fig. 6B) also demonstrated the com-ound LPE-1 hindered the EC-109 and TE-1 cells migration throughhe biological membrane. The matrigel offered a simulant biological

inued)

stroma, and EC-109 and TE-1 were blocked to invade through it in aconcentration-dependent manner after treatment with compoundLPE-1.

We also examined the expression of the typical proteins of EMTprocess. Fig. 6C showed that compound LPE-1 could up-regulatethe epithelial cells’ biomarkers, E-Cadherin and Occludin, while themesenchymal cells’ biomarkers, N-Cadherin and Vimentin, weredown-regulated correspondingly. Then the upstream transcriptionfactors, Snail and Slug, were also down-regulated after treatmentwith compound LPE-1, indicating that the compound LPE-1 mightregulate the biomarkers of EMT process through inhibiting theexpression of the transcription factors and cause the interferenceof EMT process ultimately. EC-109 cells with down-regulated LSD1expression levels by siRNA also showed upregulated the epithelialcells’ biomarkers, E-Cadherein and Occludin, while down-regulated

the mesenchymal cells’ biomarkers, N-Cadherin and Vimentin(Fig. 6D).

74 B. Wang et al. / Pharmacological Research 122 (2017) 66–77

Fig. 6. Compound LPE-1 blocked EMT process in EC-109 and TE-1 cells. (A) Wound healing assay. (B) The migration and invasion ability of EC-109 and TE-1 cells aftertreatment of compound LPE-1 was reflected using transwell and matrigel-coated transwell; (C) Expressions of Snail-1, Slug, N-Cadherin, Vimentin, E-Cadherin, Occludinwere determined after 48 h treatment of compound LPE-1. (D) Expression levels of LSD1, N-Cadherin, Vimentin, E-Cadherin, and Occludin were determined after LSD1knockdown with siRNA for 48 h. *P < 0.05 and **P 0.01 were considered statistically significant compared with the control. Dates are mean ± SD. All experiments were carriedout at least three times.

B. Wang et al. / Pharmacological Research 122 (2017) 66–77 75

(Cont

3

EcctwBwt

4

hm

Fig. 6.

.6. In vivo antitumor study

Since the obvious inhibitory activity of compound LPE-1 againstC-109 cell, we also evaluated the in vivo antitumor effect ofompound LPE-1 on xenograft model bearing EC-109 cells by sub-utaneous implantation. After the treatment of compound LPE-1,he body weight of mice, the tumor weight and the tumor volumeere measured and recorded every three days. As shown in Fig. 7A,, and D, compound LPE-1 inhibited tumor growth remarkably,hile the body weight was almost unchanged (Fig. 7C), suggesting

he antitumor efficacy and low global toxicity.

. Discussion

Esophageal cancer, the eighth most common cancer worldwide,as poor prognosis, the 5-year survival rate for esophageal squa-ous cell carcinoma (ESCC) is less than 20% [27,28]. Therefore,

inued)

novel prognostic and molecular targets for therapeutic interven-tion are urgently needed. Many researches have reported that LSD1is highly expressed in several types of cancer, such as estrogen-receptor-negative breast cancer [8,29,30], prostate cancer [26,31],bladder cancer [17], colorectal cancer [16], gastric cancer [13,14],lung cancer [9] and others. Meanwhile, recent work suggestedthat inactivation of LSD1 by small-molecule inhibitors or siRNAmay have anticancer therapeutic potential [12–14,23,26,29,32–34].Therefore, researchers have devoted to testifying the role of LSD1in ESCC [10–12,35].

From our in-house small-molecule library, compound LPE-1was identified as a new LSD1 inhibitor, which inhibited LSD1 at sub-micromolar levels (IC50 = 0.336 �M). Compound LPE-1 selectively

suppressed growth of LSD1 overexpressed EC-109 cells, while forEC-9706, TE-1, and Het-1A cells with low LSD1 expression levels,less cytotoxic effect was observed. Compound LPE-1 induced apo-ptosis, arrested the cell cycle at G2/M phase against EC109, and

76 B. Wang et al. / Pharmacological Research 122 (2017) 66–77

Fig. 7. In vivo antitumor effects of compound LPE-1. (A) Representative tumor size after treatment with compound LPE-1. Tumor weight (B), body weight (C), tumor volume( ally si

empHdeHie

i(amifhetbEcgac

ptacca

D) of the animals with the indicated treatment. **P < 0.01 were considered statistic

xhibited promising antitumor efficacy toward EC-109 xenograftodel. Additionally, after treatment of EC-109 cells with com-

ound LPE-1, the methylation levels of LSD1 substrates (H3K4me1,3K4me2 and H3K9me2) were up-regulated in a concentration-ependent manner, while the methylation level of H3K4me3 andxpression of total LSD1 were unchanged compared to H3 and4. There data suggested that the compound LPE-1 specifically

nhibited LSD1 activity at the cellular level and did not change thexpression of LSD1.

The increased motility and invasiveness of tumor cells are rem-niscent of the processes at the epithelial-mesenchymal transitionEMT), which contributes to the normal morphogenetic events suchs embryonic development, tissue remodeling, wound healing andetastasis, but also the malignance of cancer cells. The EMT process

n cell differentiation and behavior is mediated by key transcriptionactors, including Snail, zine-finger E-box-binding (ZEB) and basicelix-loop-helix transcription factors. Their expression is activatedarly in EMT process, and they further define the EMT transcrip-ion program and drive EMT procession [36–39]. Recently, it haseen reported [25] that the transcription factor Snail repressed-cadherin expression by recruiting LSD1 to its promoter, whichontributed to cancer cell invasion. Additionally, Bruno Calabretta’sroup [40] found that blocking Snail-LSD1 interaction by Parnate,n enzymatic inhibitor of LSD1, could suppress the invasiveness ofancer cells.

Therefore, we hypothesized that the potent LSD1 inhibitor, com-ound LPE-1, might suppress the migration and invasion of EC-109hrough blocking the EMT process. The data of some intuitionisticssays, such as wound healing assay, transwell and matrigel-

oated transwell assay, demonstrated that the compound LPE-1ould obviously repress the activity of migration and invasiongainst EC-109. Further investigation showed that compound LPE-

gnificant compared with the controls.

1 could up-regulate the epithelial cells’ biomarkers, E-Cadherin andOccludin, while the mesenchymal cells’ biomarkers, N-Cadherinand Vimentin, were down-regulated.

Interestingly, the upstream transcription factors, Snail and Slug,were also down-regulated after treatment with compound LPE-1.It has been recognized that LSD1 is essential for Snail1-mediatedtranscriptional repression and for maintenance of the silencedstate of Snail target genes in invasive cancer cells. The N-terminalSNAG domain of Snail1 and the amine oxidase domain of LSD1were required for their interaction [37]. Moreover, blocking theinteractions between LSD1 and Snail/Slug inhibits cancer cell inva-sion [40]. Therefore, the different observations may be related tounknown mechanisms, which require further investigations. Col-lectively, these results suggest that compound LPE-1 could reversethe EMT process, which might contribute to its antitumor effectin vitro and in vivo on EC-109 cell line and xenograft mice models.

In conclusion, our study identified a series of compounds basedon pyrimidine-thiourea hybrid and its analog LPE-1 as new potentLSD1 inhibitor. Especially, compound LPE-1 displayed obvious anti-cancer activity in vitro and in vivo against LSD1 overexpressedEC-109. Further investigations demonstrate that compound LPE-1regulated the activity of LSD1 and then reversed the EMT pro-cess, which finally induced the inhibition of migration and invasionagainst EC-109. Collectively, we identify that LSD1 may be a poten-tial therapeutic target in ESCC, and the compound LPE-1 could bea lead compound for further anti-ESCC drug discovery.

Conflict of interest

The authors declare no conflicts of interest.

gical R

A

rD2t(s

A

i0

R

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[40] G. Ferrari-Amorotti, V. Fragliasso, R. Esteki, Z. Prudente, A.R. Soliera, S.Cattelani, G. Manzotti, G. Grisendi, M. Dominici, M. Pieraccioli, G. Raschella, C.Chiodoni, M.P. Colombo, B. Calabretta, Inhibiting interactions of lysine

B. Wang et al. / Pharmacolo

cknowledgments

We are grateful for the financial support from the Natu-al Science Foundation of China (No. 21372206, 81430085); theevelopment Foundation of Priority, Ministry of Education (No.0134101130001); and National Key Research Programs of Pro-eins (2016YFA0501800); Key Research Program of Henan ProvinceNo. 161100310100); the Starting Grant of Zhengzhou Univer-ity(32210533)

ppendix A. Supplementary data

Supplementary data associated with this article can be found,n the online version, at http://dx.doi.org/10.1016/j.phrs.2017.05.25.

eferences

[1] T. Kouzarides, Chromatin modifications and their function, Cell 128 (4) (2007)693–705.

[2] C.L. Peterson, M.-A. Laniel, Histones and histone modifications, Curr. Biol. 14(14) (2004) R546–R551.

[3] Y. Shi, F. Lan, C. Matson, P. Mulligan, J.R. Whetstine, P.A. Cole, R.A. Casero, Y.Shi, Histone demethylation mediated by the nuclear amine oxidase homologLSD1, Cell 119 (7) (2004) 941–953.

[4] J. Wang, S. Hevi, J.K. Kurash, H. Lei, F. Gay, J. Bajko, H. Su, W. Sun, H. Chang, G.Xu, F. Gaudet, E. Li, T. Chen, The lysine demethylase LSD1 (KDM1) is requiredfor maintenance of global DNA methylation, Nat. Genet. 41 (1) (2009)125–129.

[5] H. Kontaki, I. Talianidis, Lysine methylation regulates E2F1-Induced celldeath, Mol. Cell 39 (1) (2010) 152–160.

[6] J. Huang, R. Sengupta, A.B. Espejo, M.G. Lee, J.A. Dorsey, M. Richter, S. Opravil,R. Shiekhattar, M.T. Bedford, T. Jenuwein, S.L. Berger, p53 is regulated by thelysine demethylase LSD1, Nature 449 (7158) (2007) 105–108.

[7] S.K.T. Ooi, C. Qiu, E. Bernstein, K. Li, D. Jia, Z. Yang, H. Erdjument-Bromage, P.Tempst, S.-P. Lin, C.D. Allis, X. Cheng, T.H. Bestor, DNMT3L connectsunmethylated lysine 4 of histone H3 to de novo methylation of DNA, Nature448 (7154) (2007) 714–717.

[8] S. Lim, A. Janzer, A. Becker, A. Zimmer, R. Schüle, R. Buettner, J. Kirfel,Lysine-specific demethylase 1 (LSD1) is highly expressed in ER-negativebreast cancers and a biomarker predicting aggressive biology, Carcinogenesis31 (3) (2010) 512–520.

[9] T. Lv, D. Yuan, X. Miao, Y. Lv, P. Zhan, X. Shen, Y. Song, Over-Expression ofLSD1 promotes proliferation, migration and invasion in non-Small cell lungcancer, PLoS One 7 (4) (2012) e35065.

10] Y. Yu, B. Wang, K. Zhang, Z. Lei, Y. Guo, H. Xiao, J. Wang, L. Fan, C. Lan, Y. Wei,Q. Ma, L. Lin, C. Mao, X. Yang, X. Chen, Y. Li, Y. Bai, D. Chen, High expression oflysine-specific demethylase 1 correlates with poor prognosis of patients withesophageal squamous cell carcinoma, Biochem. Biophys. Res. Commun. 437(2) (2013) 192–198.

11] L. Chen, Y. Xu, B. Xu, H. Deng, X. Zheng, C. Wu, J. Jiang, Over-expression oflysine-specific demethylase 1 predicts tumor progression and poor prognosisin human esophageal cancer, Int. J. Clin. Exp. Path. 7 (12) (2014) 8929–8934.

12] I. Hoshino, Y. Akutsu, K. Murakami, N. Akanuma, Y. Isozaki, T. Maruyama, T.Toyozumi, Y. Matsumoto, H. Suito, M. Takahashi, N. Sekino, A. Komatsu, T.Suzuki, H. Matsubara, Histone demethylase LSD1 inhibitors prevent cellgrowth by regulating gene expression in esophageal squamous cell carcinomacells, Ann. Surg. Oncol. 23 (1) (2016) 312–320.

13] Y.C. Zheng, Y.C. Duan, J.L. Ma, R.M. Xu, X. Zi, W.L. Lv, M.M. Wang, X.W. Ye, S.Zhu, D. Mobley, Y.Y. Zhu, J.W. Wang, J.F. Li, Z.R. Wang, W. Zhao, H.M. Liu,Triazole-dithiocarbamate based selective lysine specific demethylase 1(LSD1) inactivators inhibit gastric cancer cell growth, invasion, and migration,J. Med. Chem. 56 (21) (2013) 8543–8560.

14] L.Y. Ma, Y.C. Zheng, S.Q. Wang, B. Wang, Z.R. Wang, L.P. Pang, M. Zhang, J.W.Wang, L. Ding, J. Li, C. Wang, B. Hu, Y. Liu, X.D. Zhang, J.J. Wang, Z.J. Wang, W.Zhao, H.M. Liu, synthesis, and structure-activity relationship of novel LSD1inhibitors based on pyrimidine-thiourea hybrids as potent, orally activeantitumor agents, J. Med. Chem. 58 (4) (2015) 1705–1716.

15] E. Metzger, M. Wissmann, N. Yin, J.M. Muller, R. Schneider, A.H.F.M. Peters, T.Gunther, R. Buettner, R. Schule, LSD1 demethylates repressive histone marks

to promote androgen-receptor-dependent transcription, Nature 437 (7057)(2005) 436–439.

16] J. Ding, Z.M. Zhang, Y. Xia, G.Q. Liao, Y. Pan, S. Liu, Y. Zhang, Z.S. Yan,LSD1-mediated epigenetic modification contributes to proliferation andmetastasis of colon cancer, Br. J. Cancer 109 (4) (2013) 994–1003.

esearch 122 (2017) 66–77 77

17] W. Lan, D. Zhang, J. Jiang, The roles of LSD1-mediated epigeneticmodifications in maintaining the pluripotency of bladder cancer stem cells,Med. Hypotheses 81 (5) (2013) 823–825.

18] Y.-C. Zheng, D.-D. Shen, M. Ren, X.-Q. Liu, Z.-R. Wang, Y. Liu, Q.-N. Zhang, L.-J.Zhao, L.-J. Zhao, J.-L. Ma, B. Yu, H.-M. Liu, Baicalin, a natural LSD1 inhibitor,Bioorg. Chem. 69 (2016) 129–131.

19] S. Wang, L.-J. Zhao, Y.-C. Zheng, D.-D. Shen, E.-F. Miao, X.-P. Qiao, L.-J. Zhao, Y.Liu, R. Huang, B. Yu, H.-M. Liu, Design synthesis and biological evaluation of[1,2,4]triazolo[1,5-a]pyrimidines as potent lysine specific demethylase 1(LSD1/KDM1A) inhibitors, Eur. J. Med. Chem. 125 (2017) 940–951.

20] X.-W. Ye, Y.-C. Zheng, Y.-C. Duan, M.-M. Wang, B. Yu, J.-L. Ren, J.-L. Ma, E.Zhang, H.-M. Liu, Synthesis and biological evaluation ofcoumarin-1,2,3-triazole-dithiocarbamate hybrids as potent LSD1 inhibitors,MedChemComm 5 (5) (2014) 650–654.

21] Y.-C. Zheng, B. Yu, Z.-S. Chen, Y. Liu, H.-M. Liu, TCPs: privileged scaffolds foridentifying potent LSD1 inhibitors for cancer therapy, Epigenomics 8 (5)(2016) 651–666.

22] B. Yu, S.-Q. Wang, P.-P. Qi, D.-X. Yang, K. Tang, H.-M. Liu, Design and synthesisof isatin/triazole conjugates that induce apoptosis and inhibit migration ofMGC-803 cells, Eur. J. Med. Chem. 124 (2016) 350–360.

23] Y.C. Zheng, J. Ma, Z. Wang, J. Li, B. Jiang, W. Zhou, X. Shi, X. Wang, W. Zhao,H.M. Liu, A systematic review of histone lysine-Specific demethylase 1 and itsinhibitors, Med. Res. Rev. 5 (5) (2015) 1032–1071.

24] B. Yu, P.-P. Qi, X.-J. Shi, R. Huang, H. Guo, Y.-C. Zheng, D.-Q. Yu, H.-M. Liu,Efficient synthesis of new antiproliferative steroidal hybrids using themolecular hybridization approach, Eur. J. Med. Chem. 117 (2016) 241–255.

25] T. Lin, A. Ponn, X. Hu, B.K. Law, J. Lu, Requirement of the histone demethylaseLSD1 in Snai1-mediated transcriptional repression duringepithelial-mesenchymal transition, Oncogene 29 (35) (2010) 4896–4904.

26] M. Wang, X. Liu, J. Guo, X. Weng, G. Jiang, Z. Wang, L. He, Inhibition of LSD1 byPargyline inhibited process of EMT and delayed progression of prostatecancer in vivo, Biochem. Biophys. Res. Commun. 467 (2) (2015) 310–315.

27] P.C. Enzinger, R.J. Mayer, Esophageal cancer, New Engl. J. Med. 349 (23) (2003)2241–2252.

28] A.K. Rustgi, H.B. El-Serag, Esophageal carcinoma, New Engl. J. Med. 371 (26)(2014) 2499–2509.

29] J.A. Pollock, M.D. Larrea, J.S. Jasper, D.P. McDonnell, D.G. McCafferty,Lysine-Specific histone demethylase 1 inhibitors control Breast cancerproliferation in ER�-Dependent and –Independent manners, ACS Chem. Biol.7 (7) (2012) 1221–1231.

30] Q. Li, L. Shi, B. Gui, W. Yu, J. Wang, D. Zhang, X. Han, Z. Yao, Y. Shang, Bindingof the JmjC demethylase JARID1B to LSD1/NuRD suppresses angiogenesis andmetastasis in breast cancer cells by repressing chemokine CCL14, Cancer Res.71 (21) (2011) 6899–6908.

31] Z.-Y. Chen, H. Chen, T. Qiu, X.-D. Weng, J. Guo, L. Wang, X.-H. Liu, Effects ofcisplatin on the LSD1-mediated invasion and metastasis of prostate cancercells, Mol. Med. Rep. 14 (3) (2016) 2511–2517.

32] T. Suzuki, N. Miyata, Lysine demethylases inhibitors, J. Med. Chem. 54 (24)(2011) 8236–8250.

33] J.T. Lynch, W.J. Harris, T.C.P. Somervaille, LSD1 inhibition: a therapeuticstrategy in cancer? Expert Opin. Ther. Targets 16 (12) (2012) 1239–1249.

34] B. Dulla, K.T. Kirla, V. Rathore, G.S. Deora, S. Kavela, S. Maddika, K. Chatti, O.Reiser, J. Iqbal, M. Pal, Synthesis and evaluation of 3-amino/guanidinesubstituted phenyl oxazoles as a novel class of LSD1 inhibitors withanti-proliferative properties, Organic Biomol. Chem. 11 (19) (2013)3103–3107.

35] C. Chen, M. Zhao, N. Yin, B. He, B. Wang, Y. Yuan, F. Yu, J. Hu, B. Yin, Q. Lu,Abnormal histone acetylation and methylation levels in esophagealsquamous cell carcinomas, Cancer Invest. 29 (8) (2011) 548–556.

36] S. Lamouille, J. Xu, R. Derynck, Molecular mechanisms ofepithelial-mesenchymal transition, Nat. Rev. Mol. Cell Biol. 15 (3) (2014)178–196.

37] Y. Lin, Y. Wu, J. Li, C. Dong, X. Ye, Y.I. Chi, B.M. Evers, B.P. Zhou, The SNAGdomain of Snail1 functions as a molecular hook for recruiting lysine-specificdemethylase 1, EMBO J. 29 (11) (2010) 1803–1816.

38] H. Luo, A.K. Shenoy, X. Li, Y. Jin, L. Jin, Q. Cai, M. Tang, Y. Liu, H. Chen, D.Reisman, L. Wu, E. Seto, Y. Qiu, Y. Dou, R.A. Casero Jr., J. Lu, MOF acetylates thehistone demethylase LSD1 to suppress epithelial-to-Mesenchymal transition,Cell Rep. 15 (12) (2016) 2665–2678.

39] O.G. McDonald, H. Wu, W. Timp, A. Doi, A.P. Feinberg, Genome-scaleepigenetic reprogramming during epithelial-to-mesenchymal transition, Nat.Struct. Mol. Biol. 18 (8) (2011) 867–874.

demethylase LSD1 with snail/slug blocks cancer cell invasion, Cancer Res. 73(1) (2013) 235–245.