Novel Histone Deacetylase Inhibitors Induce Growth Arrest, Apoptosis, and Differentiation in Sarcoma Cancer Stem Cells

Novel Histone Deacetylase Inhibitors Induce Growth Arrest,Apoptosis, and Differentiation in Sarcoma Cancer Stem CellsGemma Di Pompo,†,‡,# Manuela Salerno,†,‡,# Dante Rotili,§ Sergio Valente,§ Clemens Zwergel,§

Sofia Avnet,† Giovanna Lattanzi,∥ Nicola Baldini,*,†,‡ and Antonello Mai*,§,⊥

†Orthopaedic Pathophysiology and Regenerative Medicine Unit, Istituto Ortopedico Rizzoli (IOR), 40136 Bologna, Italy‡Department of Biomedical and Neuromotor Sciences, University of Bologna, 40126 Bologna, Italy§Department of Drug Chemistry and Technologies, Sapienza University of Roma, P.le A. Moro 5, 00185 Roma, Italy∥Institute of Molecular Genetics, Unit of Bologna IOR, National Research Council of Italy, 40136 Bologna, Italy⊥Pasteur InstituteCenci Bolognetti Foundation, Sapienza University of Roma, P.le A. Moro 5, 00185 Roma, Italy

*S Supporting Information

ABSTRACT: Musculoskeletal sarcomas are aggressive malignan-cies of bone and soft tissues often affecting children andadolescents. Histone deacetylase inhibitors (HDACi) have beenproposed to counteract cancer stem cells (CSCs) in solidneoplasms. When tested in human osteosarcoma, rhabdomyosar-coma, and Ewing’s sarcoma stem cells, the new HDACi MC1742(1) and MC2625 (2) increased acetyl-H3 and acetyl-tubulin levelsand inhibited CSC growth by apoptosis induction. At nontoxicdoses, 1 promoted osteogenic differentiation. Further investigationwith 1 will be done in preclinical sarcoma models.

■ INTRODUCTION

Musculoskeletal sarcomas are relatively rare, aggressivemalignancies of bone and soft tissues often affecting childrenand young adults with devastating consequences both in termsof morbidity and mortality.1 In recent years, a uniquepopulation of cells, referred as cancer stem cells (CSCs),have been retained responsible for cancer development andprogression, response to therapy, and metastatis in differentsolid tumors,2 including sarcomas.3,4 Lysine acetylation anddeacetylation of histone and nonhistone proteins is a crucialevent in the epigenetic modulation of gene expression, cellcycle progression, and signal transduction cascades.5,6 Novelperspectives for the targeting of CSCs within sarcomas mayderive from the employment of HDACi. Some authorsprovided evidence of the effectiveness of these compoundsagainst CSCs derived from different tumors. Specifically, it hasbeen reported that HDACi induce differentiation of breastCSCs,7 promote apoptosis and cell cycle arrest in CSCs of headand neck squamous cell carcinoma,8 and exert strongantiproliferative effects on glioblastoma and colorectalcarcinoma CSCs.9,10 However, to date, no information abouttheir activity on CSCs derived from human sarcomas isavailable. Nevertheless, some data indicate that thesecompounds are effective in inhibiting the growth of nativesarcoma cells, as demonstrated by several preclinical stud-ies.11−13 This is possibly due to the involvement of histone

acetylation in the differentiation process of mesenchymalprecursors,14 the hypothetical cell-of-origin of transformedsarcoma cells.15 Therefore, we investigated the effects ofdifferent HDACi in CSC cultures derived from human modelsof osteosarcoma (OS), rhabdomyosarcoma (RMS), andEwing’s sarcoma (ES). We used MC1742 (1, Figure 1A),belonging to the series of the uracil-based hydroxyamides(UBHAs),16,17 MC2625 (2, Figure 1B), the aza-analogue of

Received: January 21, 2015

Figure 1. Design of the novel HDACi 1 and 2 and structure of theSIRTi 3 tested in this study.

Brief Article

pubs.acs.org/jmc

© XXXX American Chemical Society A DOI: 10.1021/acs.jmedchem.5b00126J. Med. Chem. XXXX, XXX, XXX−XXX

aroylammino cinnamyl hydroxamates previously described byus,19−22 and MC2141 (3, Figure 1C), a sirtuin (class IIIHDAC) inhibitor active in the low micromolar range againstSIRT1 (IC50 = 9.8 μM) and SIRT2 (IC50 = 12.3 μM). Theclassical UBHA scaffold16 in 1 was modified by converting theC6-phenyl into a C6-(4-biphenyl) moiety, a structural changewhich improved the anti-HDAC activity of UBHAs17 as well asother hydroxamates.18,19 Compound 2 displays a 2,3-diphenylpropanamide chain, related to some branched groupshighly efficient in improving potency and selectivity in HDACinhibition.22,23 The SIRTi 3 was described as able to induceantiproliferative effects in different cancer cell lines includinghuman glioblastoma and colorectal carcinoma CSCs.10

■ CHEMISTRYThe synthetic routes followed for the preparation of 1 and 2 aredepicted in Scheme 1. The hydroxamate 1 was synthesized

from the 5-((4-([1,1′-biphenyl]-4-yl)-6-oxo-1,6-dihydropyrimi-din-2-yl)thio) pentanoic acid 417 by treatment with (i) ethylchloroformate and triethylamine, (ii) O-(2-methoxy-2-propyl)hydroxylamine, and (iii) Amberlyst 15 ion-exchange resin inmethanol at room temperature (Scheme 1A). For the synthesisof 2, 2-bromo-5-aminopyridine was treated with 2,3-diphenyl-propanoyl chloride (previously prepared reacting 2,3-diphenyl-propanoic acid with thionyl chloride at 80 °C) and triethyl-amine in anhydrous dichlomethane to furnish N-(6-bromopyr-idin-3-yl)-2,3-diphenylpropanamide 5. Afterward, theintermediate 5 underwent Heck coupling reaction using butylacrylate, tetra-n-butyl ammonium iodide, sodium acetatetrihydrate, triphenylphosphine, and palladium acetate inanhydrous N,N-dimethylformamide at 140 °C overnight toafford the ester 6. By hydrolysis of 6, the acid 7 was obtainedand converted to the hydroxamate 2 through the methodalready described for 1 (Scheme 1B).

■ RESULTS AND DISCUSSION

HDAC Inhibitory Enzyme Assays. The two novel HDACi1 and 2 were tested against all the HDAC1−11 isoforms in 10-dose IC50 mode with 3-fold serial dilution starting from 50 μMsolutions (Table 1). From these data, 1 emerged as a class I/IIb-selective HDACi, being potent at submicromolar/nano-molar level against class I HDACs (HDAC1−3, 8) and atnanomolar level against class IIb HDACs (HDAC6, 10).Against class IIa HDACs (HDAC4, 5, 7, 9), 1 did not showappreciable inhibition at 50 μM, the maximum tested dose.Differently, 2 behaved as a pan-HDACi, being efficient ininhibiting all the 11 HDAC isoforms in the range 0.08−12 μM.In particular, it showed the highest nanomolar inhibitorypotency against HDAC3 and HDAC6, inhibited HDAC8 atsubmicromolar concentration, and it remained still activeagainst class IIa HDACs with IC50 values around 10 μM.Thus, both of them deserved to be investigated in sarcomaCSCs, 1 as more potent and selective HDACi, 2 as a broaderinhibitor including class IIa HDACs in its spectrum of action.In addition, we included 310 in our studies as representative of aspecific class III HDAC (sirtuin) inhibitor.

Screening of Cytotoxic Effects of 1−3 in SarcomaCSCs. The cytotoxicity of the investigated HDACi wasmeasured on CSCs generated from three different histotypesof sarcoma: RMS (RD and A204 cell lines), OS (MG-63 andHOS cell lines), and ES (SK-ES-1 and A673 cell lines). Thedrug concentrations were arbitrarily chosen starting from a highconcentration (100 μM) and with dilutions until the lowestconcentration (0.1 μM). Compound cytoxicity was evaluated incomparison to cells treated with DMSO alone in order toexclude any nonspecific cytotoxicity caused by the vehicle. After72 h of treatment, compounds exerted different effects on CSCcultures (see the relative curves and half-maximal inhibitorconcentration (IC50) values in Supporting Information (SI),Figure S1; p = 0.0209). 3 was effective on CSCs, although inalmost all cultures the effect was observed starting from 10 μM.Adversely, 1 and 2 were able to significantly affect cell viabilityeven at 0.1 and 1 μM, depending on the cell line, showing aclear dose-dependent activity and the lowest IC50 values foreach CSC culture, calculated by linear regression (SI, FigureS1). To dissect the role of specific HDAC class inhibition insarcoma CSCs’ cytotoxicity, we tested also SAHA,24 the pan-HDACi approved by FDA in 2006 for the treatment of CTCL,MS-275, a class I-selective HDACi,25 and MC1568,26−30 a classII-selective HDACi, on RD, MG-63, and SK-ES-1 cells. Theresults and the IC50 values are shown in SI, Figure S2 (p =0.0209; for MC1568, data not shown).

Antiproliferative Effects of 1 and 2 on Sarcoma CSCs.On the basis of the MTT results, 1 and 2 showed the best IC50values and the highest cytotoxic activity. For this reason, theirantiproliferative effect on sarcospheres was confirmed bydetermining the number of viable cells by a growth curve. Amore restricted range of concentrations (2, 1, and 0.5 μM) wasemployed on the base of IC50 values calculated for each cellhistotype. As shown in Figure 2, all the concentrations of both

Scheme 1a

a(a) (i) ClCOOEt, (Et)3N, THF, 0 °C, (ii) NH2OC(Me)2OMe, rt,(iii) Amberlyst 15, MeOH, rt; (b) (i) SOCl2, 80 °C, (ii) Et3N,anhydrous CH2Cl2, 0 °C; (c) P(Ph)3, Pd(OAc)2, n-Bu4NI, AcONa·3H2O, butyl acrylate, anhydrous DMF, 140 °C; (d) 2N KOH, EtOH/H2O, rt.

Table 1. IC50 Values (μM) of 1 and 2 against the HDAC1-11 Isoforms

IC50 vs HDAC (μM)

compd 1 2 3 4 5 6 7 8 9 10 11

1 0.10 0.11 0.02 >50 >50 0.007 >50 0.61 >50 0.04 0.12 1.42 1.77 0.08 11.7 9.37 0.01 8.77 0.61 10.6 1.8 10.2

Journal of Medicinal Chemistry Brief Article

DOI: 10.1021/acs.jmedchem.5b00126J. Med. Chem. XXXX, XXX, XXX−XXX

B

drugs reduced cell proliferation in all CSC cultures in a dose-dependent manner. In particular, 1 significantly inhibited cellproliferation already after 48 h, with the exception of 0.5 μMfor A673 (Figure 2A; p < 0.05), showing the highesteffectiveness. The treatment with the lowest dose of 2 for 48h significantly affected MG-63, RD, and SK-ES-1 viability, whileall CSCs were sensitive at 1 and 2 μM (Figure 2B; p < 0.05).Nevertheless, all the 1 and 2 concentrations exerted significanteffects on all the CSC cultures after 72 h (Figure 2; p < 0.05).Inhibition of HDAC Activity in Sarcoma CSCs.

Immunofluorescence and Western blotting were carried outto demonstrate the inhibitory activity of the selected HDACion deacetylation of histone H3. MG-63 CSCs have beenchosen among all the CSC cultures as a representative modelbecause it resulted as being the most sensitive to the HDACiantiproliferative effect. The 2 and 0.5 μM concentrations wereselected because they were the highest and the lowest used inthe growth assay, respectively. After 24 h of treatment with 1and 2, a dose-dependent increase of acetylation, observable aspunctuate nuclear staining of acetyl-histone H3, was detectedby immunofluorescence analysis (Figure 3A). Accordingly,protein level of acetyl-histone H3, detected by Western blot(WB), was enhanced in HDACi-treated CSCs in respect tountreated cells (Figure 3B, left). The densitometric analysisshowed that 1 was significantly effective both at 0.5 and 2 μM,while 2 was effective only at 2 μM (Figure 3B, right; p =0.0495), in agreement with their relative degrees of enzymeinhibition. When tested with antiacetyl-tubulin antibodies byWB as functional test for HDAC6 inhibition, again 1 was morepotent than 2 in increasing acetyl-tubulin levels (Figure 3B; p =0.0495).Apoptosis Induction in Sarcoma CSCs. The Hoechst

33258 staining was used to verify if the selected compoundswere able to affect sarcoma CSC growth through apoptosis. Asshown by representative pictures of MG-63 CSCs, 1 and 2generated an increase of the presence of apoptotic cells withconcentrated dense granular fluorescence compared to un-treated cells, especially at 1 and 2 μM, whereas no relevantapoptosis was detectable in control cells (Figure 4A).The quantification of cells with apoptotic bodies revealed a

dose-dependent increase of apoptosis in all CSC cultures after48 h of incubation with HDACi, consistent with the effect on

cell proliferation. Treatment with 1 and 2 μM of 1 and 2significantly induced apoptosis of all CSC cultures, with theexception of A204 CSCs, that were less sensitive to treatmentwith 2 (Figure 4B; *p < 0.05, **p < 0.01). In MG-63 CSCs, theinduction of apoptosis by 1 and 2 was demonstrated byevaluating the presence of the molecular marker phosphatidylserine (PS) on the surface of the cytoplasmic membrane(Figure S3 in SI).

Osteogenic Differentiation Induction in SarcomaCSCs. Differentiation therapy has been proposed as apromising therapeutic strategy to revert chemoresistance ofCSCs in different tumor histotypes.31,32 To determine whetherdifferentiation occurred following the 1 and 2 cell treatment,Alizarin Red S staining was carried out on MG-63 CSCs,selected as representative OS model. Alizarin Red S selectivelybinds to calcium salts, therefore the rate of mineralized noduleformation is directly correlated to the amount of eluted dye,detected by spectrophotometry. Because cytotoxicity ofcompounds could negatively influence osteogenic differ-

Figure 2. Antiproliferative effect of 1 and 2 on different sarcoma CSCcultures after 24, 48, and 72 h of treatment. The antitumor activity of 1(A) and 2 (B) was determined by growth curves on sarcoma CSCsafter treatment with 0.5, 1, and 2 μM of compounds. *p < 0.05 vscontrol.

Figure 3. Inhibition of histone acetylation after HDACi treatment. (A)Immunofluorescence staining of acetyl-histone H3 (acetyl K9) inuntreated cells (control, upper panel) or treated with 1 (middle panel)and 2 (lower panel) at different concentrations (0.5 or 2 μM). Nucleiwere counterstained with DAPI. The merged images are shown in theleft column. Representative images, scale bar 10 μm. (B) Westernblotting of acetyl-H3 and acetyl-tubulin in MG-63 CSCs treated with0.5 and 2 μM of 1 and 2 for 24 h (left, representative image) anddensitometric analysis (right) reported as index vs negative control(bold line). *p < 0.05 vs control. Normalization to TBP.

Journal of Medicinal Chemistry Brief Article

DOI: 10.1021/acs.jmedchem.5b00126J. Med. Chem. XXXX, XXX, XXX−XXX

C

entiation and consequently cell ability to mineralize, nontoxicconcentrations of drugs were selected on the basis of IC50values previously calculated. The Alamar Blue test, performedat the mineralization end-point, confirmed that only the highestconcentration of 2 significantly inhibited cell growth (Figure S4in SI; p < 0.05). The elution of the Alizarin Red S dye indicatedthat 2 treatment did not improve mineralization with respect tocontrol cells. On the contrary, treatment with 1, starting fromthe lowest dose (0.025 μM) up to the highest (0.5 μM),successfully enhanced bone nodule formation in a significantdose-dependent manner (Figure 5; p < 0.05).

■ CONCLUSIONSHere we show the effects of three HDACi, 1 (class I/IIb-selective), 2 (nonselective), and 3 (class III-selective) on CSCsgenerated from three different histotypes of human sarcomas(RMS, OS, and ES). The first screening and the resultant IC50values demonstrated a potent cytotoxic effect by the “classical”HDACi 1 and 2 that reduced cell viability in some CSCcultures even at the lowest tested concentration, 0.1 μM, whilethe sirtuin inhibitor 3 was less effective. In the same test, SAHA(a pan-HDACi) showed comparable (MG-63 cells) or lower(RD and SK-ES-1 cells) potency than 1, the class-I selective

MS-275 was less effective, and the class II-selective MC1568inhibited the tested CSCs at 38.3 (MG-63), 1,733 (RD), and50.7 (SK-ES-1) μM (data not shown), thus suggesting thatconcurrent class I and IIb HDACs inhibition is crucial to obtainanticancer effects in sarcoma CSCs. On these bases, theantiproliferative activities of 1 and 2 against sarcospheres on thesix tested CSCs were determined. To demonstrate the HDACinhibition activity of 1 and 2 in cells, we detected an increase ofthe acetyl-H3 and acetyl-tubulin levels after exposure of MG-63CSCs, chosen as the most sensitive to the selected HDACi, toboth compounds. Notably, 1 resulted more effective than 2,also at the lowest concentration, confirming its strongantideacetylase activity. About apoptosis induction in sarco-spheres, both 1 and 2 significantly enhanced the number ofapoptotic cells and induced PS as a molecular marker, the firstbeing again the most effective. We ultimately evaluated theosteogenic differentiation induction of OS CSCs after exposureto nontoxic concentrations of 1 and 2. As expected, 1 promotedhigh cell differentiation evaluated in terms of mineral nodulesformation in vitro, whereas 2 was ineffective, probably due to itslower anti-HDAC potency.In conclusion, this study provides an adequate background

about the in vitro efficacy of a novel uracil-based HDACi, 1, ona panel of CSCs from different human sarcoma histotypes, thussupporting the use of HDACi-based therapy also in theseneoplasms.Further investigation is required to clarify its activity in in

vivo preclinical sarcoma models.

■ EXPERIMENTAL SECTIONChemistry.Melting point were determined on a Buchi 530 melting

point apparatus and are uncorrected. 1H NMR spectra were recordedat 400 MHz on a Bruker AC 400 spectrometer; reporting chemicalshifts in δ (ppm) units relative to the internal referencetetramethylsilane (Me4Si). All compounds were routinely checkedby TLC and 1H NMR. TLC was performed on aluminum-backed silicagel plates (Merck DC, Alufolien Kieselgel 60 F254) with spotsvisualized by UV light. Yields of all reactions refer to the purifiedproducts. All chemicals were purchased from Aldrich Chimica, Milan(Italy), and were of the highest purity. Mass spectra were recorded ona API-TOF Mariner by Perspective Biosystem (Stratford, TX, USA),and samples were injected by an Harvard pump using a flow rate of 5−10 μL/min, infused in the Electrospray system. Elemental analyseswere obtained by a PE 2400 (Perkin- Elmer) analyzer and have beenused to determine purity of the described compounds, that is, >95%.Analytical results are within ±0.40% of the theoretical values.

Figure 4. Apoptosis induction after HDACi treatment. Presence ofapoptotic CSCs after 48 h of exposure to 0.5, 1, and 2 μM of 1 and 2.(A) Hoechst 33258 nuclear staining of MG-63 CSCs, showing thebasal level of cells with apoptotic bodies in untreated control (left) anda dose-dependent increase in the number of apoptotic cells aftertreatment with 1 (middle) and 2 (right). Representative pictures, scalebar 50 μm. (B) Quantification of the percentage of apoptotic cells overdifferent fields in OS (top), RMS (middle), and ES (bottom) CSCsevaluated after Hoechst 33258 staining. *p < 0.05 and **p < 0.01 vscontrol.

Figure 5. Mineralization assay after HDACi treatment. MG-63 CSCswere treated with nontoxic doses of 1 and 2 under differentiatingconditions for 14 days. Quantification of Alizarin Red S stainingevaluated by dye elution with cetylpyridinium chloride, showing asignificant dose-dependent induction of mineral nodules formationafter 1 treatment. *p < 0.05 vs control.

Journal of Medicinal Chemistry Brief Article

DOI: 10.1021/acs.jmedchem.5b00126J. Med. Chem. XXXX, XXX, XXX−XXX

D

Synthesis of 5-((4-([1,1′-Biphenyl]-4-yl)-6-oxo-1,6-dihydropyrimi-din-2-yl)thio)-N-hydroxypentanamide (1). To a 0 °C cooled solutionof 417 (0.53 mmol, 200 mg) in dry tetrahydrofuran (4 mL), ethylchloroformate (1.3 mmol, 138 mg, 0.12 mL) and triethylamine (1.38mmol, 140 mg, 0.19 mL) were added and the mixture was stirred for10 min. The solid was filtered off, and O-(2-methoxy-2-propyl)-hydroxylamine (3.18 mmol, 0.24 mL) was added to the filtrate. Theresulting mixture was stirred at room temperature (rt) for 1 h then wasevaporated under reduced pressure, and the residue was diluted inMeOH (2.5 mL). Amberlyst 15 ion-exchange resin (106 mg) wasadded to the solution of the O-protected hydroxamate, and themixture was stirred at rt for 1 h. Afterward, the reaction was filteredand the filtrate was concentrated in vacuum to give the crude 1, whichwas purified by recrystallization; mp 202−204 °C; yield 48%.Recrystallization system: methanol. 1H NMR (DMSO-d6) δ 1.69(m, 4H, CH2CH2CH2CH2S), 2.02 (t, 2H, CH2CO), 3.25 (t, 2H,CH2S), 6.72 (s, 1H, C5-H), 7.39−7.48 (m, 3H, benzene rings), 7.73−7.83 (m, 4H, benzene rings), 8.14 (m, 2H, benzene rings), 8.69 (s, 1H,NHOH), 10.37 (s, 1H, NHOH), 12.6 (s, 1H, uracil NH). 13C NMR(DMSO-d6) δ 24.6, 31.7, 32.9, 36.1, 112.4, 127.1 (2C), 127.6, 127.9(2C), 128.0 (2C), 129.2 (2C), 134.8, 140.8, 140.9, 160.6, 163.9, 167.7,169.6 ppm. Anal. (C21H21N3O3S) % Calcd: C, 63.78; H, 5.35; N,10.63; S, 8.11. Found (%) : C, 64.04; H, 5.40; N, 10.49; S, 8.01. MS(ESI), m/z: 396 [M + H].Synthesis of N-(2-Bromopyridin-5-yl)-2,3-diphenyl Propanamide

(5). Triethylamine (4.73 mmoli, 0.66 mL) and 2,3-diphenylpropanoylchloride (3.93 mmol, 0.96 g, previously prepared by reaction of thecorresponding acid with thionyl chloride (5 mL) at 80 °C for 1 h)were added to a solution of 2-bromo-5-aminopyridine (3.93 mmol,0.68 g) in anhydrous dichloromethane (10 mL) cooled at 0 °C. After1.5 h, the reaction was quenched by water (50 mL) and extracted withdichloromethane (3 × 50 mL). The collected organic phases werewashed with saturated sodium chloride solution (100 mL), dried, andconcentrated to obtain a solid residue that was purified by silica gelchromatography eluting with ethyl acetate/chloroform 1/1 to affordpure 5 as a colorless solid recrystallized by toluene; mp 131−133 °C;yield 85%. 1H NMR (CDCl3) δ 3.02−3.07 (dd, 1H, PhCHHCO−),3.55−3.61 (dd, 1H, PhCHHCO−), 3.80−3.84 (t, 1H, PhCH2−),7.12−7.32 (m, 11H, pyridine benzene and −NHCO− protons), 7.91−7.94 (m, 2H, pyridine protons), 8.00 (s, 1H, pyridine proton) ppm.13C NMR (CDCl3) δ 47.5, 48.4, 126.2, 127.7, 128.5 (2C), 129.0, 129.5(2C), 129.8 (2C), 130.3, 130.7 (2C), 136.9, 138.4, 139.5, 141.9, 145.3,173.0 ppm. MS (ESI), m/z 381 [M + H]+.Synthesis of Butyl 3-(5-(2,3-Diphenylpropanamido)pyridin-2-

yl)acrylate (6). Triphenylphosphine (0.55 mmol, 0.143 g) andpalladium acetate (0.27 mmol, 0.06 g) were added under nitrogenatmosphere to a solution of 5 (6.56 mmol, 2.50 g), tetra-n-butylammonium iodide (6.56 mmol, 2.42 g), sodium acetate trihydrate(17.05 mmol, 2.32 g), water (0.7 mL), and butyl acrylate (13.11 mmol,2.0 mL) in anhydrous N,N-dimethylformamide (13.0 mL) and insealed tube. The resulting mixture was stirred at 140 °C overnight,then the reaction was quenched by water (50 mL) and extracted withethyl acetate (3 × 50 mL). The collected organic phases were washedwith saturated sodium chloride solution (100 mL), dried, andconcentrated to obtain an oily residue that was purified by silica gelchromatography eluting with ethyl acetate/n-hexane 1:2 to providepure 6 as a pale-yellow oil; yield 42%. 1H NMR (CDCl3) δ 0.95−0.99( t , 3H , OCH 2CH 2CH 2CH 3 ) , 1 . 4 1− 1 . 4 7 (m , 2H ,OCH2CH2CH2CH3), 1.65−1.71 (m, 2H, OCH2CH2CH2CH3),3.06−3.11 (dd, 1H, PhCHHCO−), 3.59−3.65 (dd, 1H,PhCHHCO−), 3.78−3.81 (t, 1H, PhCH2−), 4.20−4.23 (t, 2H,OCH2CH2CH2CH3), 6.77−6.81 (d, 1H, CH2CH2COOBu), 7.11−7.38 (m, 12H, benzene protons, pyridine proton and NHCO), 7.59−7.63 (d d, 1H, CH2CH2COOBu), 8.15−8.17 (d, 1H, pyridineproton), 8.33−8.34 (d, 1H, pyridine proton) ppm. 13C NMR (CDCl3)δ 13.9, 18.5, 31.5, 47.7, 48.4, 65.0, 116.5, 121.3, 125.9, 127.9 (2C),128.7 (2C), 129.4 (2C), 129.8 (2C), 130.7 (2C), 138.5, 139.8 (2C),140.6, 144.8, 149.0, 166.7, 172.7 ppm. MS (ESI), m/z: 429 [M + H]+.Synthesis of 3-(5-(2,3-Diphenylpropanamido)pyridin-2-yl)acrylic

Acid (7). A solution of 2N potassium hydroxide (2.52 mmol, 0.14 g)

was added to a solution of butyl 3-(5-(2,3-diphenylpropanamido)-pyridin-2-yl)acrylate 6 (1.26 mmol, 0.54 g) in ethanol (5 mL), and thefinal solution was stirred at rt overnight. Then a 2 N HCl solution (5mL) was added dropwise to the reaction, and the precipitated solidwas filtered, washed with water (3 × 10 mL) and dried to give the pure7 as a colorless solid recrystallized by toluene; mp 123−125 °C; yield83%. 1H NMR (DMSO-d6) δ 2.99−3.02 (dd, 1H, PhCHHCO−),3.40−3.45 (dd, 1H, PhCHHCO−), 4.04−4.07 (t, 1H, PhCH2−),6.66−6.70 (d, 1H, CH2CH2COOH), 7.15−7.45 (m, 11H, CH2CH2COOH and benzene protons), 7.59−7.61 (d, 1H, pyridineproton), 8.04−8.06 (d, 1H, pyridine proton), 8.70 (s, 1H, pyridineproton), 10.51 (bs, 1H, NHCO), 12.00 (bs, 1H, COOH) ppm. 13CNMR (DMSO-d6) δ 47.4, 48.5, 120.8, 122.9, 126.2, 127.7 (2C), 128.7(2C), 129.5 (2C), 129.8 (2C), 130.6 (2C), 138.3, 139.5, 140.3, 142.5,144.9, 148.9, 171.6, 172.7 ppm. MS (ESI), m/z: 371 [M − H]−.

Synthesis of 3-(5-(2,3-Diphenylpropanamido)pyridin-2-yl)-N-hy-droxyacrylamide (2). To a cooled (0 °C) solution of 3-(5-(2,3-diphenylpropanamido)pyridin-2-yl)acrylic acid 7 (0.32 mmol, 0.12 g)in anhydrous THF (5 mL), ethyl chloroformate (0.35 mmol, 0.03 mL)and triethylamine (0.39 mmol, 0.05 mL) were added and the resultingmixture was stirred at 0 °C for 15 min. The solid was filtered andwashed with anhydrous THF (3 × 5 mL), and then O-(2-methoxy-2-propyl)hydroxylamine (0.97 mmol, 0.07 mL) was added to thesolution at 0 °C and stirred at rt for 1 h. After this time, the solventwas removed under vacuum, the residue was eluted with methanol (5mL), and Amberlist 15 ion-exchange resin (0.032 g) was added to thissolution. The resulting mixture was stirred for 1 h, then the resin wasfiltered and the solution concentrated to give 2 that was recrystallizedby acetonitrile; mp 178−180 °C; yield 68%. 1H NMR (DMSO-d6) δ2.98−3.03 (dd, 1H, PhCHHCO−), 3.40−3.45 (dd, 1H,PhCHHCO−), 4.06−4.10 (t, 1H, PhCH2−), 6.78−6.82 (d, 1H,CH2CH2CONHOH), 7.15−7.46 (m, 11H, benzene protons),7.55−7.57 (d, 1H, pyridine proton), 8.06−8.08 (d, 1H, pyridineproton), 8.74 (s, 1H, pyridine proton), 9.5 (bs, 1H, NHOH), 10.84(bs, 1H, NHCO), 11.0 (bs, 1H, NHOH) ppm. 13C NMR (DMSO-d6)δ 47.4, 48.5, 120.8, 122.9, 126.2, 127.7 (2C), 128.7 (2C), 129.5 (2C),129.8 (2C), 130.6 (2C), 138.3, 139.5, 140.3, 142.5, 144.9, 148.9, 161.8,172.7 ppm. Anal. (C23H21N3O3) % calcd: C, 71.30; H, 5.46; N, 10.85.Found (%): C, 71.19; H, 5.41; N, 11.04. MS (ESI), m/z: 388 [M +H]+.

HDAC1−11 Isoforms Inhibition Assay. See SI.Biology: CSC Cultures. See SI.Cell Viability Assay. See SI.Cell Growth Assay. Single cells obtained after dissociation of

sarcospheres were plated in 12-well plates (50000/well) in completeIMDM in order to induce cell adhesion. After 24 h, cells wereincubated with three concentrations (0.5, 1, and 2 μM) of 1 and 2 thatwere selected on the basis of their IC50. After 24, 48, and 72 h, thenumber of viable cells was evaluated by dye exclusion viability assay.The percentage of growth inhibition was calculated in respect tonegative control cells (exposed to DMSO). The experiment wasrepeated three times in duplicate.

Immunofluorescence. To confirm the activity of the HDACi inCSCs, the acetylation level of histone H3 was detected byimmunofluorescence. Briefly, single MG-63 cells obtained afterdissociation of sarcospheres were seeded on glass coverslips incomplete IMDM in order to induce cell adhesion and then exposed to0.5 and 2 μM of 1 and 2. After 10 h, the cells were fixed in 4%paraformaldehyde, permeabilized in 0.15% Triton X-100, saturatedusing 4% BSA in PBS at rt for 1 h, and incubated overnight withantiacetyl-histone H3K9 (Abcam, Cambridge, UK) at 1:50 dilution inPBS-4% BSA and then with a FITC-conjugated secondary antibody.Slides were mounted in glycerol-DABCO and observed using a NikonEclipse E600 microscope equipped with a digital camera. Images wereelaborated only for brightness and contrast using Adobe Photoshop 7.

Western Blotting. See SI.Apoptosis Analysis. Dissociated sarcospheres were seeded on

glass coverslips in complete IMDM in order to induce cell adhesionand then exposed to 0.5, 1, and 2 μM of 1 and 2. Untreated cells wereconsidered as negative control. After 48 h, cells were fixed by

Journal of Medicinal Chemistry Brief Article

DOI: 10.1021/acs.jmedchem.5b00126J. Med. Chem. XXXX, XXX, XXX−XXX

E

methanol/acetic acid (3:1) for 5 min, hydrated with 7% NaHCO31:200 in Hank’s balanced salt solution (HBSS) at rt for 5 min, anddark incubated in 2.25 μg/mL of Hoechst 33258 (Sigma-Aldrich) at rtfor 10 min. Cells were observed under fluorescence microscopy todetermine the number cells with apoptotic nuclear bodies over at leastfive different fields at 20× objective. Results were expressed aspercentage of apoptotic cells over the fields.Osteogenic Differentiation. Osteogenic differentiation was

evaluated by in vitro assessing of cell mineralization ability of oneselected CSC culture. The 10000 cells derived from MG-63sarcospheres were seeded in α-MEM (Sigma-Aldrich) supplementedwith 10% FBS (complete α-MEM) in order to induce cell adhesion.After reaching a confluence of 70% cells, mineralization was inducedby mineralization medium (MM medium), consisting of complete α-MEM supplemented with 10 mM β-glycerophosphate, 10−8 Mdexamethasone, and 50 mg/mL L-ascorbic acid 2-phosphate (Sigma-Aldrich). During the differentiation period, the cells were continuouslyexposed to subcytotoxic concentrations of 1 and 2 (0.5, 0.1, 0.05, and0.025 μM). After 14 days, cell viability was assessed by Alamar Blueassay (see SI). Then, the cells were stained by Alizarin Red S, whichselectively binds to calcium salts, in order to evaluate the ability of cellto deposit mineral nodules. Briefly, cells were fixed with 3.7%paraformaldehyde at rt for 20 min, washed with PBS, and stained with2% Alizarin Red S (Sigma-Aldrich), pH 4.2, in distilled H2O at rt for 1h. The cells were washed thoroughly several times with distilled H2Oto remove unspecific staining. To quantify the content of calcium salts,Alizarin Red dye was eluted by incubation with 10% cetylpyridiniumchloride in distilled H2O at rt for 15 min under gentle shaking. Theabsorbance was measured at 570 nm with a microplate reader (TecanInfinite F200pro, Crailsheim, Germany). Data were expressed as ODmeasured in treated sample in respect to negative control (MMmedium alone). The experiment was performed in triplicate.Statistical Analysis. See SI.

■ ASSOCIATED CONTENT*S Supporting InformationExperimental for HDAC assay, CSC cultures, cell viabilityassay, Western blotting, annexin V-FICT assay, Alamar Blueassay, and statistical analysis. The Supporting Information isavailable free of charge on the ACS Publications website atDOI: 10.1021/acs.jmedchem.5b00126.

■ AUTHOR INFORMATIONCorresponding Authors*For N.B.: phone, +39 051 6366678; fax, +39 051 6366974; E-mail, [email protected].*For A.M.: phone, +39 06 49913392; fax, +39 06 49693268; E-mail, [email protected] Contributions#G. Di Pompo and M. Salerno share first authorship of thisarticle.NotesThe authors declare no competing financial interest.

■ ACKNOWLEDGMENTSWe thank Garland R. Marshall, Department of Biochemistryand Molecular Biophysics, Washington University School ofMedicine, St. Louis, Missouri, USA, for critical reading of themanuscript and scientific discussion. This work was supportedby FIRB grants (RBAP10447J to N.B. and RBFR10ZJQT toA.M.) from the Italian Ministry of Education, University andResearch, by the “5 per mille” 2010 (N.B.) and RF-2010-2318330 (A.M.) grant from the Italian Ministry of the Health,Financial Support for Scientific Research, and by the ItalianAssociation for Cancer Research (AIRC 11426 to N.B.),

Sapienza Ateneo Project 2013 (D.R.), PRIN 2012 (prot.2012C-TAYSY) (D.R.), IIT-Sapienza Project (A.M.), FP7 ProjectsBLUEPRINT/282510 and A-PARADDISE/602080 (A.M.).

■ ABBREVIATIONS USED

CSCs, cancer stem cells; CTCL, cutaneous T-cell lymphoma;ES, Ewing’s sarcoma; FBS, fetal bovine serum; FICT-,fluorescein isothiocyanate-; IMDM, Iscove’s Modified Dulbec-co’s Medium; α-MEM, α-minimum essential medium; OS,osteosarcoma; RMS, rhabdomyosarcoma; UBHAs, uracil-basedhydroxyamides

■ REFERENCES(1) Montero Luis, A.; Perez Aguilar, D.; Lopez Martín, J. A.Multidisciplinary management of soft tissue sarcomas. Clin. Transl.Oncol. 2010, 12, 543−553.(2) Jordan, C. T.; Guzman, M. L.; Noble, M. Cancer stem cells. N.Engl. J. Med. 2006, 355, 1253−1261.(3) Gibbs, C. P.; Kukekov, V. G.; Reith, J. D.; Tchigrinova, O.;Suslov, O. N.; Scott, E. W.; Ghivizzani, S. C.; Ignatova, T. N.;Steindler, D. A. Stem-like cells in bone sarcomas: implications fortumorigenesis. Neoplasia 2005, 7, 967−976.(4) Salerno, M.; Avnet, S.; Bonuccelli, G.; Eramo, A.; De Maria, R.;Gambarotti, M.; Gamberi, G.; Baldini, N. Sphere-forming cell subsetswith cancer stem cell properties in human musculoskeletal sarcomas.Int. J. Oncol. 2013, 43, 95−102.(5) Mai, A.; Massa, S.; Rotili, D.; Cerbara, I.; Valente, S.; Pezzi, R.;Simeoni, S.; Ragno, R. Histone deacetylation in epigenetics: anattractive target for anticancer therapy. Med. Res. Rev. 2005, 25, 261−309.(6) Falkenberg, K. J.; Johnstone, R. W. Histone deacetylases and theirinhibitors in cancer, neurological diseases and immune disorders.Nature Rev. Drug Discovery 2014, 13, 673−691.(7) Salvador, M. A.; Wicinski, J.; Cabaud, O.; Toiron, Y.; Finetti, P.;Josselin, E. The histone deacetylase inhibitor abexinostat inducescancer stem cells differentiation in breast cancer with low Xistexpression. Clin. Cancer Res. 2013, 19, 6520−6531.(8) Chikamatsu, K.; Ishii, H.; Murata, T.; Sakakura, K.; Shino, M.;Toyoda, M.; Takahashi, K.; Masuyama, K. Alteration of cancer stemcell-like phenotype by histone deacetylase inhibitors in squamous cellcarcinoma of the head and neck. Cancer Sci. 2013, 104, 1468−1475.(9) Rotili, D.; Tarantino, D.; Nebbioso, A.; Paolini, C.; Huidobro, C.;Lara, E.; Mellini, P.; Lenoci, A.; Pezzi, R.; Botta, G.; Lahtela-Kakkonen,M.; Poso, A.; Steinkuhler, C.; Gallinari, P.; De Maria, R.; Fraga, M.;Esteller, M.; Altucci, L.; Mai, A. Discovery of salermide-related sirtuininhibitors: binding mode studies and antiproliferative effects in cancercells including cancer stem cells. J. Med. Chem. 2012, 55, 10937−10947.(10) Rotili, D.; Tarantino, D.; Carafa, V.; Paolini, C.; Schemies, J.;Jung, M.; Botta, G.; Di Maro, S.; Novellino, E.; Steinkuhler, C.; DeMaria, R.; Gallinari, P.; Altucci, L.; Mai, A. Benzodeazaoxaflavins assirtuin inhibitors with antiproliferative properties in cancer stem cells.J. Med. Chem. 2012, 55, 8193−8197.(11) Wu, Z.; Ma, C.; Shan, Z.; Ju, Y.; Li, S.; Zhao, Q. Histonedeacetylase inhibitors suppress the growth of human osteosarcomas invitro and in vivo. J. BUON 2013, 18, 1032−1037.(12) Kutko, M. C.; Glick, R. D.; Butler, L. M.; Coffey, D. C.; Rifkind,R. A.; Marks, P. A.; Richon, V. M.; La Quaglia, M. P. Histonedeacetylase inhibitors induce growth suppression and cell death inhuman rhabdomyosarcoma in vitro. Clin. Cancer Res. 2003, 9, 5749−5755.(13) Liu, S.; Cheng, H.; Kwan, W.; Lubieniecka, J. M.; Nielsen, T. O.Histone deacetylase inhibitors induce growth arrest, apoptosis, anddifferentiation in clear cell sarcoma models. Mol. Cancer Ther. 2008, 7,1751−1761.(14) Siddiqi, S.; Mills, J.; Matushansky, I. Epigenetic remodeling ofchromatin architecture: exploring tumor differentiation therapies in

Journal of Medicinal Chemistry Brief Article

DOI: 10.1021/acs.jmedchem.5b00126J. Med. Chem. XXXX, XXX, XXX−XXX

F

mesenchymal stem cells and sarcomas. Curr. Stem Cell Res. Ther. 2010,5, 63−73.(15) Xiao, W.; Mohseny, A. B.; Hogendoorn, P. C.; Cleton-Jansen, A.M. Mesenchymal stem cell transformation and sarcoma genesis. Clin.Sarcoma Res. 2013, 3, 10.(16) Mai, A.; Massa, S.; Rotili, D.; Simeoni, S.; Ragno, R.; Botta, G.;Nebbioso, A.; Miceli, M.; Altucci, L.; Brosch, G. Synthesis andbiological properties of novel, uracil-containing histone deacetylaseinhibitors. J. Med. Chem. 2006, 49, 6046−6056.(17) Mai, A.; Perrone, A.; Nebbioso, A.; Rotili, D.; Valente, S.;Tardugno, M.; Massa, S.; De Bellis, F.; Altucci, L. Novel uracil-based 2-aminoanilide and 2-aminoanilide-like derivatives: histone deacetylaseinhibition and in-cell activities. Bioorg. Med. Chem. Lett. 2008, 18,2530−2535.(18) Mai, A.; Massa, S.; Valente, S.; Simeoni, S.; Ragno, R.; Bottoni,P.; Scatena, R.; Brosch, G. Aroyl-pyrrolyl hydroxyamides: influence ofpyrrole c4-phenylacetyl substitution on histone deacetylase inhibition.ChemMedChem 2006, 1, 225−237.(19) Mai, A.; Massa, S.; Pezzi, R.; Valente, S.; Loidl, P.; Brosch, G.Synthesis and biological evaluation of 2-, 3-, and 4-acylaminocinnamyl-N-hydroxyamides as novel synthetic HDAC inhibitors. Med. Chem.2005, 1, 245−254.(20) Thaler, F.; Varasi, M.; Colombo, A.; Boggio, R.; Munari, D.;Regalia, N.; Rozio, M. G.; Reali, V.; Resconi, A. E.; Mai, A.; Gagliardi,S.; Dondio, G.; Minucci, S.; Mercurio, C. Synthesis and biologicalcharacterization of amidopropenyl hydroxamates as HDAC inhibitors.ChemMedChem 2010, 5, 1359−1372.(21) Thaler, F.; Colombo, A.; Mai, A.; Amici, R.; Bigogno, C.;Boggio, R.; Cappa, A.; Carrara, S.; Cataudella, T.; Fusar, F.; Gianti, E.;di Ventimiglia, S. J.; Moroni, M.; Munari, D.; Pain, G.; Regalia, N.;Sartori, L.; Vultaggio, S.; Dondio, G.; Gagliardi, S.; Minucci, S.;Mercurio, C.; Varasi, M. Synthesis and biological evaluation of N-hydroxyphenylacrylamides and N-hydroxypyridin-2-ylacrylamides asnovel histone deacetylase inhibitors. J. Med. Chem. 2010, 53, 822−839.(22) Valente, S.; Tardugno, M.; Conte, M.; Cirilli, R.; Perrone, A.;Ragno, R.; Simeoni, S.; Tramontano, A.; Massa, S.; Nebbioso, A.;Miceli, M.; Franci, G.; Brosch, G.; Altucci, L.; Mai, A. Novel cinnamylhydroxyamides and 2-aminoanilides as histone deacetylase inhibitors:apoptotic induction and cytodifferentiation activity. ChemMedChem2011, 6, 698−712.(23) Yang, F.; Zhang, T.; Wu, H.; Yang, Y.; Liu, N.; Chen, A.; Li, Q.;Li, J.; Qin, L.; Jiang, B.; Wang, X.; Pang, X.; Yi, Z.; Liu, M.; Chen, Y.Design and optimization of novel hydroxamate-based histonedeacetylase inhibitors of bis-substituted aromatic amides bearingpotent activities against tumor growth and metastasis. J. Med. Chem.2014, 57, 9357−9369.(24) Marks, P. A.; Breslow, R. Dimethyl sulfoxide to vorinostat:development of this histone deacetylase inhibitor as an anticancerdrug. Nature Biotechnol. 2007, 25, 84−90.(25) Knipstein, J.; Gore, L. Entinostat for treatment of solid tumorsand hematologic malignancies. Expert Opin. Invest. Drugs 2011, 20,1455−1467.(26) Mai, A.; Massa, S.; Pezzi, R.; Simeoni, S.; Rotili, D.; Nebbioso,A.; Scognamiglio, A.; Altucci, L.; Loidl, P.; Brosch, G. Class II (IIa)-selective histone deacetylase inhibitors. 1. Synthesis and biologicalevaluation of novel (aryloxopropenyl)pyrrolyl hydroxamides. J. Med.Chem. 2005, 48, 3344−3353.(27) Nebbioso, A.; Manzo, F.; Miceli, M.; Conte, M.; Manente, L.;Baldi, A.; De Luca, A.; Rotili, D.; Valente, S.; Mai, A.; Usiello, A.;Gronemeyer, H.; Altucci, L. Selective class II HDAC inhibitors impairmyogenesis by modulating the stability and activity of HDAC-MEF2complexes. EMBO Rep. 2009, 10, 776−782.(28) Spallotta, F.; Rosati, J.; Straino, S.; Nanni, S.; Grasselli, A.;Ambrosino, V.; Rotili, D.; Valente, S.; Farsetti, A.; Mai, A.; Capogrossi,M. C.; Gaetano, C.; Illi, B. Nitric oxide determines mesodermicdifferentiation of mouse embryonic stem cells by activating class IIahistone deacetylases: potential therapeutic implications in a mousemodel of hindlimb ischemia. Stem Cells 2010, 28, 431−442.

(29) Colussi, C.; Rosati, J.; Straino, S.; Spallotta, F.; Berni, R.; Stilli,D.; Rossi, S.; Musso, E.; Macchi, E.; Mai, A.; Sbardella, G.; Castellano,S.; Chimenti, C.; Frustaci, A.; Nebbioso, A.; Altucci, L.; Capogrossi, M.C.; Gaetano, C. Nε-Lysine acetylation determines dissociation fromGAP junctions and lateralization of connexin 43 in normal anddystrophic heart. Proc. Natl. Acad. Sci. U. S. A. 2011, 108, 2795−2800.(30) Lenoir, O.; Flosseau, K.; Ma, F. X.; Blondeau, B.; Mai, A.;Bassel-Duby, R.; Ravassard, P.; Olson, E. N.; Haumaitre, C.;Scharfmann, R. Specific control of pancreatic endocrine beta- anddelta-cell mass by class IIa histone deacetylases HDAC4, HDAC5, andHDAC9. Diabetes 2011, 60, 2861−2871.(31) Azzi, S.; Bruno, S.; Giron-Michel, J.; Clay, D.; Devocelle, A.;Croce, M.; Ferrini, S.; Chouaib, S.; Vazquez, A.; Charpentier, B.;Camussi, G.; Azzarone, B.; Eid, P. Differentiation therapy: targetinghuman renal cancer stem cells with interleukin 15. J. Natl. Cancer Inst.2011, 103, 1884−1898.(32) Lombardo, Y.; Scopelliti, A.; Cammareri, P.; Todaro, M.; Iovino,F.; Ricci-Vitiani, L.; Gulotta, G.; Dieli, F.; De Maria, R.; Stassi, G. Bonemorphogenetic protein 4 induces differentiation of colorectal cancerstem cells and increases their response to chemotherapy in mice.Gastroenterology 2011, 140, 297−309.

Journal of Medicinal Chemistry Brief Article

DOI: 10.1021/acs.jmedchem.5b00126J. Med. Chem. XXXX, XXX, XXX−XXX

G