PHASCI 3120

No. of Pages 10, Model 5G

15 October 2014 European Journal of Pharmaceutical Sciences xxx (2014) xxx–xxx 1

Contents lists available at ScienceDirect

European Journal of Pharmaceutical Sciences journal homepage: www.elsevier.com/locate/ejps 6 7 3 4

Q1

5 8

Q2

9 10 11 12 13 14 15 16 17

Joana Soares a,b, Nuno A.L. Pereira d, Ângelo Monteiro d, Mariana Leão a,b, Cláudia Bessa a,b, Daniel J.V.A. dos Santos d,e, Liliana Raimundo a,b, Glória Queiroz a,c, Alessandra Bisio f, Alberto Inga f, Clara Pereira a,b, Maria M.M. Santos d,⇑, Lucília Saraiva a,b,⇑ a

REQUIMTE, Faculdade de Farmácia, Universidade do Porto, Rua de Jorge Viterbo Ferreira n.° 164, 4050-313 Porto, Portugal Laboratório de Microbiologia, Departamento de Ciências Biológicas, Faculdade de Farmácia, Universidade do Porto, Rua de Jorge Viterbo Ferreira n.° 164, 4050-313 Porto, Portugal c Laboratório de Farmacologia, Departamento de Ciências do Medicamento, Faculdade de Farmácia, Universidade do Porto, Rua de Jorge Viterbo Ferreira n.° 164, 4050-313 Porto, Portugal d Research Institute for Medicines (iMed.ULisboa), Faculty of Pharmacy, Universidade de Lisboa, Av. Prof. Gama Pinto, 1649-003 Lisboa, Portugal e REQUIMTE, Departamento de Química e Bioquímica, Faculdade de Ciências, Universidade do Porto, Rua do Campo Alegre, 4169-007 Porto, Portugal f CIBIO, Centre for Integrative Biology, Laboratory of Transcriptional Networks, University of Trento, Via delle Regole, 101, 38123 Trento, Italy b

Q3

18 19

a r t i c l e

2 6 1 3 22 23 24 25 26 27 28 29 30 31 32 33 34 35

Oxazoloisoindolinones with in vitro antitumor activity selectively activate a p53-pathway through potential inhibition of the p53–MDM2 interaction

i n f o

Article history: Received 6 August 2014 Received in revised form 29 September 2014 Accepted 2 October 2014 Available online xxxx

Q4

Keywords: Antitumor activity Docking analysis p53 p53–MDM2 interaction Oxazoloisoindolinones Yeast-based assay

a b s t r a c t One of the most appealing targets for anticancer treatment is the p53 tumor suppressor protein. In half of human cancers, this protein is inactivated due to endogenous negative regulators such as MDM2. Actually, restoring the p53 activity, particularly through the inhibition of its interaction with MDM2, is considered a valuable therapeutic strategy against cancers with a wild-type p53 status. In this work, we report the synthesis of nine enantiopure phenylalaninol-derived oxazolopyrrolidone lactams and the evaluation of their biological effects as p53–MDM2 interaction inhibitors. Using a yeast-based screening assay, two oxazoloisoindolinones, compounds 1b and 3a, were identified as potential p53–MDM2 interaction inhibitors. The molecular mechanism of oxazoloisoindolinone 3a was further validated in human colon adenocarcinoma HCT116 cells with wild-type p53 (HCT116 p53+/+) and in its isogenic derivative without p53 (HCT116 p53 / ). Indeed, using these cells, we demonstrated that oxazoloisoindolinone 3a exhibited a p53-dependent in vitro antitumor activity through induction of G0/G1-phase cell cycle arrest and apoptosis. The selective activation of a p53-apoptotic pathway by oxazoloisoindolinone 3a was further supported by the occurrence of PARP cleavage only in p53-expressing HCT116 cells. Moreover, oxazoloisoindolinone 3a led to p53 protein stabilization and to the up-regulation of p53 transcriptional activity with increased expression levels of several p53 target genes, as p21, MDM2, BAX and PUMA, in p53+/+ but not in p53 / HCT116 cells. Additionally, the ability of oxazoloisoindolinone 3a to block the p53–MDM2 interaction in HCT116 p53+/+ cells was confirmed by co-immunoprecipitation. Finally, the molecular docking analysis of the interactions between the synthesized compounds and MDM2 revealed that oxazoloisoindolinone 3a binds to MDM2. Altogether, this work adds, for the first time, the oxazoloisoindolinone scaffold to the list of chemotypes activators of a wild-type p53-pathway with promising antitumor activity. Moreover, it may open the way to the development of a new class of p53–MDM2 interaction inhibitors. Ó 2014 Published by Elsevier B.V.

37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61

62

⇑ Corresponding authors at: REQUIMTE, Laboratório de Microbiologia, Departamento de Ciências Biológicas, Faculdade de Farmácia, Universidade do Porto, Rua de Jorge Viterbo Ferreira n.° 164, 4050-313 Porto, Portugal. Tel.: +351 220428584; fax: +351 226093390 (L. Saraiva). Tel.: +351 217946451; fax: +351 217946470 (M.M.M. Santos). E-mail addresses: [email protected] (M.M.M. Santos), lucilia.saraiva@ ff.up.pt (L. Saraiva).

1. Introduction

63

The p53 tumor suppressor protein, widely known as ‘‘the guardian of the genome’’, is a key regulator of important cellular processes, such as cell cycle, DNA repair, cell death, and angiogenesis. Inactivation of p53 function is therefore a common event in human cancers. In approximately half of human cancers, the p53

64

http://dx.doi.org/10.1016/j.ejps.2014.10.006 0928-0987/Ó 2014 Published by Elsevier B.V.

Q1 Please cite this article in press as: Soares, J., et al. Oxazoloisoindolinones with in vitro antitumor activity selectively activate a p53-pathway through potential inhibition of the p53–MDM2 interaction. Eur. J. Pharm. Sci. (2014), http://dx.doi.org/10.1016/j.ejps.2014.10.006

65 66 67 68

PHASCI 3120

No. of Pages 10, Model 5G

15 October 2014 2 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113

J. Soares et al. / European Journal of Pharmaceutical Sciences xxx (2014) xxx–xxx

function is inactivated by deletion or mutation of the TP53 gene. In the remaining half of human cancers, the p53 protein retains a wild-type (wt) status, but its function is compromised by distinct mechanisms. One of the major inhibitory mechanisms of wt p53 is mediated by its interaction with the oncoprotein murine double minute 2 (MDM2). MDM2 and p53 mutually regulate each other through an auto-regulatory feedback loop. As a transcription factor, p53 increases the expression level of the MDM2 gene upon activation. In turn, MDM2 binds to the p53 transactivation domain and inhibits its transcriptional activity, favors its export out of the nucleus, and promotes its proteasome-mediated degradation. Therefore, restoration of p53 activity by inhibiting the p53– MDM2 interaction represents an appealing therapeutic strategy for tumors with wt p53. This has been the focus of a large effort Q5 in anticancer drug discovery (Lauria et al., 2011; Wang et al., 2012; Wang and Hu, 2012; Zhao et al., 2013a). In the last years, several classes of small molecule inhibitors of the p53–MDM2 interaction have been developed (Kamal et al., 2012). The cis-imidazoline compounds, termed nutlins, were the first potent and selective inhibitors of the p53–MDM2 interaction identified (e.g. Nutlin-3a) (Vassilev et al., 2004). However, this class of compounds has shown to be unsuitable for human use. An optimized member of the nutlin family with improved potency and selectivity, RG7112, is under clinical trials (Ray-Coquard et al., 2012). Other chemical families reported as potent inhibitors of the p53–MDM2 interaction are the spirooxindoles (e.g. MI-888) (Zhao et al., 2013b), the piperidinones (e.g. AM-8553) (Rew et al., 2012), and the isoindolinones (e.g. NU8231) (Hardcastle et al., 2005). Each of them detains a unique, rigid heterocyclic scaffold, from which three lipophilic groups are projected into p53 pocket in MDM2, mimicking the three pivotal PHE19, TRP23 and LEU26 residues. More classes of small molecules have emerged as inhibitors of the p53–MDM2 interaction, but only few molecules have advanced into clinical trials due to their low selectivity, high toxicity and poor pharmacokinetic profiles (Zhao et al., 2013a). Efforts are therefore in progress to identify new scaffolds of p53–MDM2 interaction inhibitors with improved pharmacological properties. These molecules have the potential to be developed as anticancer drug candidates (Lauria et al., 2011; Wang et al., 2012; Wang and Hu, 2012; Zhao et al., 2013a). In the present work, synthesis combined with computational docking studies and biological screening assays led to the identification of oxazoloisoindolinone 3a, a compound with potent in vitro antitumor activity through selective activation of a p53-pathway involving the inhibition of the p53–MDM2 interaction.

114

2. Materials and methods

115

2.1. Chemistry

116

Chemicals were purchased from Sigma–Aldrich Chemical Co. Ltd., Spain and used without further purification. All compounds were characterized by NMR spectroscopy and mass spectrometry. NMR spectra was acquired on a Bruker 400 Ultrashield spectrometer at 400 MHz (1H NMR) and 100 MHz (13C NMR). Data were reported as follows: chemical shift (d), multiplicity (s: singlet, d: doublet, dd: doublet of doublet; ddd: doublet doublet of doublets, t: triplet, td: triplet of doublets, m: multiplet, br: broad), coupling constants (J) in Hertz and integration. 1H and 13C chemical shifts are expressed in ppm using the solvent as internal reference. Elemental analysis for C, H and N elements were performed in a Flash 2000 Elemental Analyzer (ThermoScientific, UK) at Faculty of Pharmacy of Lisbon University. The purity of compounds submitted for biological testing was determined by elemental analysis to be P95%. Melting points were determined in a Bock-M Monoscope

117 118 119 120 121 122 123 124 125 126 127 128 129 130

apparatus and are uncorrected. The specific rotation values were measured in a Perkin–Elmer 241MC Polarimeter. The flash chromatography was performed using Merck Silica Gel (200–400 mesh).

131

2.1.1. General preparation of compounds 1a–c, 2a–b, 3a–b and 4a–b To a stirred solution of the appropriate aminoalcohol (0.1 g, 0.66 mmol) in toluene (10 mL), it was added 1.1 eq of the appropriate b-oxo-acid. The mixture was then refluxed under a dean–stark apparatus and in an inert atmosphere. After cooling the reaction to room temperature the solvent was evaporated to dryness. The crude residue obtained was adsorbed onto silica and purified by flash chromatography with the adequate solvent system.

135

2.1.1.1. (3S,9bR)-3-benzyl-9b-methyl-2,3-dihydrooxazolo[2,3-a]isoindol-5(9bH)-one (1a). A mixture of (S)-2-amino-3-phenyl-propan1-ol (0.1 g, 0.66 mmol) and 2-acetyl-benzoic acid (0.12 g, 0.73 mmol) in toluene (10 mL) was treated according to the general procedure for 16 h. Purification using EtOAc/n-Hex (2:3) afforded compound 1a (0.170 g, 92%) as a colorless oil. [a]20 D = +27.3 (c = 2.1, CH2Cl2). Elemental Anal. calcd. for C18H17NO2
0.75H2O: C 73.82, H 6.38, N 4.78, found: C 73.33, H 5.96, N 4.80. 1H and 13C NMR spectras were found to be identical to the ones described previously (Pereira et al., 2013).

143

2.1.1.2. (3S,9bR)-3-benzyl-9b-phenyl-2,3-dihydrooxazolo[2,3-a]isoindol-5(9bH)-one (1b). A mixture of (S)-2-amino-3-phenyl-propan1-ol (0.1 g, 0.66 mmol) and 2-benzoyl-benzoic acid (0.16 g, 0.73 mmol) in toluene (10 mL) was treated according to the general procedure for 16 h. Purification using EtOAc/n-Hex (3:7), followed by recrystallization in Et2O/n-Hex, afforded compound 1b (0.193 g, 86%) as a white crystal solid. [a]20 D = +144.6 (c = 1.9, CH2Cl2). Elemental Anal. calcd. for C23H19NO2: C 80.91, H 5.62, N 4.10, found: C 80.91, H 5.68, N 4.25. 1H and 13C NMR spectras were found to be identical to the ones described previously (Pereira et al., 2013).

153

2.1.1.3. (3S,9bR)-3-benzyl-6,7-dimethoxy-2,3-dihydrooxazolo[2,3-a]isoindol-5(9bH)-one (1c). A mixture of (S)-2-amino-3-phenyl-propan1-ol (0.1 g, 0.66 mmol) and 6-formyl-2,3-dimethoxy-benzoic acid (0.16 g, 0.73 mmol) in toluene (10 mL) was treated according to the general procedure for 16 h. Purification using EtOAc/n-Hex (1:3), followed by recrystallization in EtOAc/n-Hex, afforded compound 1c (0.141 g, 66%) as a white solid. mp 124–127 °C. [a]20 D = +32.8 (c = 1.2, CH2Cl2). 1H NMR (400 MHz, CDCl3) dH (ppm) 7.34–7.19 (m, 5H, H-ar), 7.15 (d, J = 8.1 Hz, 1H, H-ar), 7.04 (d, J = 8.1 Hz, 1H, H-ar), 5.61 (s, 1H, CH), 4.49–4.37 (m, 1H, CH), 4.30– 4.20 (m, 1H, CH2), 4.05 (s, 3H, CH3), 3.89 (dd, J = 8.6, 6.7 Hz, 1H, CH2), 3.85 (s, 3H, CH3), 3.12 (dd, J = 13.9, 5.5 Hz, 1H, CH2), 2.90 (dd, J = 13.8, 8.1 Hz, 1H, CH2). 13C NMR (101 MHz, CDCl3) dC (ppm) 171.95 (C@O), 154.29 (C-q), 147.57 (C-q), 137.02 (C-q), 135.28 (C-q), 129.55 (C-ar), 128.72 (C-ar), 126.88 (C-ar), 124.49 (C-q), 119.03 (C-ar), 116.84 (C-ar), 90.08 (CH), 74.78 (CH2), 62.52 (CH3), 56.73 (CH3), 55.77 (CH), 39.88 (CH2). IR (KBr) 1698 (C@O) cm 1. Elemental Anal. calcd. for C19H19NO4
0.15 H2O: C 69.56, H 5.94, N 4.27, found: C 69.18, H 5.59, N 3.97.

164

2.1.1.4. (3S,7aR)-3-benzyl-7a-methyltetrahydropyrrolo[2,1-b]oxazol5(6H)-one (2a). A mixture of (S)-2-amino-3-phenyl-propan-1-ol (0.1 g, 0.66 mmol) and levulinic acid (0.08 mL, 0.73 mmol) in toluene (10 mL) was treated according to the general procedure for 48 h. Purification using EtOAc/n-Hex (3:7) afforded compound 2a (0.111 g, 73%). Elemental Anal. calcd. for C14H17NO2
0.5H2O: C 69.97, H 7.57, N 5.83, found: C 69.63, H 7.40, N 5.85. 1H and 13C NMR spectras were found to be identical to the ones described in literature (Allin et al., 2001).

183

Q1 Please cite this article in press as: Soares, J., et al. Oxazoloisoindolinones with in vitro antitumor activity selectively activate a p53-pathway through potential inhibition of the p53–MDM2 interaction. Eur. J. Pharm. Sci. (2014), http://dx.doi.org/10.1016/j.ejps.2014.10.006

132 133 134

136 137 138 139 140 141 142

144 145 146 147 148 149 150 151 152

154 155 156 157 158 159 160 161 162 163

165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182

184 185 186 187 188 189 190 191

PHASCI 3120

No. of Pages 10, Model 5G

15 October 2014 J. Soares et al. / European Journal of Pharmaceutical Sciences xxx (2014) xxx–xxx 192 193 194 195 196 197 198 199 200 201

202 203 204 205 206 207 208 209 210 211 212

213 214 215 216 217 218 219 220 221 222 223

224 225 226 227 228 229 230 231 232

2.1.1.5. (3S,7aS)-3-benzyl-7a-phenyltetrahydropyrrolo[2,1-b]oxazol5(6H)-one (2b). A mixture of (S)-2-amino-3-phenyl-propan-1-ol (0.1 g, 0.66 mmol) and 3-benzoyl-propionic acid (0.13 g, 0.73 mmol) in toluene (10 mL) was treated according to the general procedure for 48 h. Purification with EtOAc/n-Hex (3:7), followed by recrystallization in Et2O/n-Hex, afforded compound 2b (0.140 g, 72%) as a white solid. [a]20 D = +4.7 (c = 1.9, CH2Cl2). Elemental Anal. calcd. for C19H19NO2: C 77.78, H 6.54, N 4.78, found: C 77.75, H 6.68, N 4.98. 1H and 13C NMR spectras were found to be identical to the ones described previously (Pereira et al., 2013). 2.1.1.6. (3R,9bS)-3-benzyl-9b-phenyl-2,3-dihydrooxazolo[2,3-a]isoindol-5(9bH)-one (3a). A mixture of (R)-2-amino-3-phenyl-propan1-ol (0.1 g, 0.66 mmol) and 2-benzoyl-benzoic acid (0.16 g, 0.73 mmol) in toluene (10 mL) was treated according to the general procedure for 16 h. Purification using EtOAc/n-Hex (1:9), followed by recrystallization in Et2O/n-Hex, afforded compound 3a (0.167 g, 74%) as a white crystal solid. [a]20 138.0 (c = 2.2, CH2Cl2). D = Elemental Anal. calcd. for C23H19NO2: C 80.91, H 5.62, N 4.10, found: C 81.05, H 5.69, N 4.26. 1H and 13C NMR spectras were found to be identical to the one obtained for compound 1b (Pereira et al., 2013). 2.1.1.7. (3R,9bS)-3-benzyl-6,7-dimethoxy-2,3-dihydrooxazolo[2,3-a]isoindol-5(9bH)-one (3b). A mixture of (R)-2-amino-3-phenyl-propan1-ol (0.1 g, 0.66 mmol) and 6-formyl-2,3-dimethoxy-benzoic acid (0.16 g, 0.73 mmol) in toluene (10 mL) was treated according to the general procedure for 16 h. Purification using EtOAc/n-Hex (1:3), followed by recrystallization in EtOAc/n-Hex, afforded compound 3b (0.122 g, 57%) as a colorless crystal solid. [a]20 30.1 D = (c = 1.2, CH2Cl2). Elemental Anal. calcd. for C19H19NO4.0.15H2O: C 69.56, H 5.94, N 4.27, found: C 69.18, H 5.59, N 3.97. The 1H NMR spectra was found to be identical to the one obtained for compound 1c. 2.1.1.8. (3R,7aS)-3-benzyl-7a-methyltetrahydropyrrolo[2,1-b]oxazol5(6H)-one (4a). A mixture of (R)-2-amino-3-phenyl-propan-1-ol (0.1 g, 0.66 mmol) and levulinic acid (0.08 mL, 0.73 mmol) in toluene (10 mL) was treated according to the general procedure for 16 h. Purification using EtOAc/n-Hex (1:2) afforded compound 4a (0.093 g, 61%) as a pale yellow solid. [a]20 52.7 (c = 0.9, D = CH2Cl2). Elemental Anal. calcd. for C14H17NO2
0.15 H2O: C 69.56, H 5.94, N 4.27, found: C 69.18, H 5.59, N 3.97. The 1H NMR spectra was found to be identical to the one obtained for compound 2a.

242

2.1.1.9. (3R,7aR)-3-benzyl-7a-phenyltetrahydropyrrolo[2,1-b]oxazol5(6H)-one (4b). A mixture of (R)-2-amino-3-phenyl-propan-1-ol (0.1 g, 0.66 mmol) and 3-benzoyl-propionic acid (0.13 g, 0.73 mmol) in toluene (10 mL) was treated according to the general procedure for 48 h. Purification with EtOAc/n-Hex (3:7), followed by recrystallization in EtOAc/n-Hex, afforded compound 4b (0.165 g, 84%) as a white solid. [a]20 3.0° (c = 2.4, CH2Cl2). D = Elemental Anal. calcd. for C19H19NO2
0.15H2O: C 77.07, H 6.58, N 4.73, found: C 77.09, H 6.42, N 4.82. The 1H NMR spectra was found to be identical to compound 2b.

243

2.2. Biological studies

244

For the yeast phenotypic assay, the yeast expression vectors pGADT7-(LEU2) encoding human MDM2 under the control of an ADH1 constitutive promoter was kindly provided by Dr. Xue-Min Zhang (from National Center of Biomedical Analysis, China); pLS89-(TRP1) encoding human wt p53 under the control of a GAL1-10 inducible promoter was kindly provided by Dr. Richard Iggo (from Swiss Institute for Experimental Cancer Research,

233 234 235 236 237 238 239 240 241

245 246 247 248 249 250

3

Switzerland). Nutlin-3a was provided from Alexis Biochemicals (Grupo Taper, Sintra, Portugal). All tested compounds were dissolved in dimethyl sulfoxide (DMSO; Sigma–Aldrich, Sintra, Portugal).

251

2.2.1. Yeast strain, transformation, and growth conditions For the yeast growth-inhibition assay, Saccharomyces cerevisiae CG379 was co-transformed using the standard lithium acetate method. For selection of co-transformed yeast, cells were routinely grown in a minimal selective medium with 2% (w/w) glucose, 0.7% (w/w) yeast nitrogen base without amino acids from Difco (Quilaban, Sintra, Portugal) and all the amino acids required for yeast growth (50 lg/mL) except leucine and tryptophan, to approximately 1 optical density at 600 nm (OD600). To induce expression of wt p53, yeast cells were diluted to 0.05 OD600 into selective induction medium with 2% (w/w) galactose and 1% (w/w) raffinose (instead of glucose), and incubated at 30 °C under continuous orbital shaking (200 r.p.m.) for approximately 42 h (time required by control yeast, co-transformed with the empty vectors pLS239 and pGADT7, to achieve 0.45 OD600). Yeast growth was analyzed by counting the number of colony-forming units (CFU) per mL after 2 days incubation at 30 °C on Sabouraud Dextrose Agar plates from Liofilchem (Frilabo, Porto, Portugal).

255

2.2.2. Effect of compounds on yeast cell growth Co-transformed yeast cells were incubated in selective induction medium in the presence of 10 lM Nutlin-3a, 1–50 lM of the tested compounds or 0.1% DMSO only, for approximately 42 h, at 30 °C, under continuous shaking. Yeast cell growth was analyzed by CFU counts as described above.

273

2.2.3. Yeast cell cycle analysis Flow cytometric analysis of DNA content was performed using Sytox Green Nucleic Acid, as described (Leão et al., 2013c). Briefly, 1  107 cells were fixed in 70% (v/v) ethanol, incubated with 250 lg/mL RNase A from Sigma–Aldrich (Sintra, Portugal) and 1 mg/mL Proteinase K from Sigma–Aldrich (Sintra, Portugal), and further incubated with 10 lM Sytox Green Nucleic Acid from Invitrogen (Alfagene, Carcavelos, Portugal), followed by flow cytometric analysis.

279

2.2.4. Human tumor cell lines and growth conditions The human colon adenocarcinoma HCT116 cell line harboring a wt p53 form (HCT116 p53+/+) and its isogenic derivative, in which p53 has been knocked out (HCT116 p53 / ) were kindly provided by Dr. B. Vogelstein (The Johns Hopkins Kimmel Cancer Center, Baltimore, Maryland, USA). Cell lines were routinely cultured in RPMI-1640 with ultraglutamine medium from Lonza (VWR, Carnaxide, Portugal) supplemented with 10% fetal bovine serum from Gibco (Alfagene, Carcavelos, Portugal) and maintained in a humidified incubator at 37 °C with 5% CO2 in air.

288

2.2.5. Effect of compounds on the in vitro growth of human tumor cells Cells were plated in 96-well plates at a final density of 5.0  103 cells/well and incubated for 24 h. Cells were then exposed to serial dilutions of each compound (from 1.85 to 150 lM). The effect of the compounds was analyzed following 48 h incubation, using the sulforhodamine B (SRB) assay. Briefly, following fixation with 10% trichloroacetic acid from Scharlau (Sigma–Aldrich, Sintra, Portugal), plates were stained with 0.4% SRB from Sigma–Aldrich (Sintra, Portugal) and washed with 1% acetic acid. The bound dye was then solubilized with 10 mM Tris Base and the absorbance was measured at 510 nm in a microplate reader (Biotek Instruments Inc., Synergy MX, USA). The solvent of the compounds (DMSO) corresponding to the maximum concentration used in

298

Q1 Please cite this article in press as: Soares, J., et al. Oxazoloisoindolinones with in vitro antitumor activity selectively activate a p53-pathway through potential inhibition of the p53–MDM2 interaction. Eur. J. Pharm. Sci. (2014), http://dx.doi.org/10.1016/j.ejps.2014.10.006

252 253 254

256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272

274 275 276 277 278

280 281 282 283 284 285 286 287

289 290 291 292 293 294 295 296 297

299 300 301 302 303 304 305 306 307 308 309 310

PHASCI 3120

No. of Pages 10, Model 5G

15 October 2014 4 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373

J. Soares et al. / European Journal of Pharmaceutical Sciences xxx (2014) xxx–xxx

these assays (0.25%) was included as control. The concentration of compound that causes a 50% reduction in the net protein increase in cells (GI50) was determined for all tested compounds. 2.2.6. Analysis of cell cycle and apoptosis in tumor cell lines HCT116 p53+/+ cells were plated in 6-well plates at a final density of 1.5  105 cells/well and incubated for 24 h. Cells were then treated with the GI50 concentration of Nutlin-3a (3.2 lM), compound 1b (8.8 lM), compound 3a (9.7 lM), with twice the GI50 concentration (2  GI50) or DMSO only (control) for 24 h. For cell cycle analysis, cells were thereafter fixed in ice-cold 70% ethanol and incubated at 37 °C with RNase A (Sigma–Aldrich, Sintra, Portugal) at final concentration of 20 lg/mL for 15 min, and further incubated with 50 lg/mL PI from Fluka (Sigma–Aldrich, Sintra, Portugal) for 30 min, followed by flow cytometric analysis. For apoptosis analysis, cells were analyzed by flow cytometry using the Annexin V: FITC Apoptosis Detection Kit I from BD Biosciences (Enzifarma, Porto, Portugal), according to the manufacturer’s instructions. 2.2.7. Western blot analysis To analyze the protein expression in HCT116 human tumor cell lines, cells were plated in 6-well plates at a final density of 1.5  105 cells/well and incubated for 24 h. Cells were then treated with the GI50 concentration of Nutlin-3a (3.2 lM), compound 1b (8.8 lM), compound 3a (9.7 lM) or DMSO only for 4, 8, 24 and 48 h. Whole cell lysates were prepared by lysing the cells with RIPA buffer (Sigma–Aldrich, Sintra, Portugal) in the presence of EDTAfree protease inhibitor cocktail (Sigma–Aldrich, Sintra, Portugal). Following whole protein quantification using the Coomassie staining (Bradford, Sigma–Aldrich, Sintra, Portugal), proteins (40 lg) were electrophoresed on 10% SDS–PAGE and transferred to a Whatman nitrocellulose membrane from Protan (VWR, Carnaxide, Portugal). Membranes were blocked with 5% milk and probed with a mouse monoclonal anti-p53 (DO-1), anti-MDM2 (D-12), anti-Bax (2D2), anti-PUMA (B-6) or anti-PARP (C2-10) antibodies from Santa Cruz Biotechnology (Frilabo, Porto, Portugal), followed by an antimouse horseradish-peroxidase (HRP)-conjugated secondary antibody from Santa Cruz Biotechnology (Frilabo, Porto, Portugal). For p21 detection, membranes were probed with a rabbit polyclonal anti-p21 antibody (C-19) from Santa Cruz Biotechnology (Frilabo, Porto, Portugal), followed by an anti-rabbit horseradishperoxidase (HRP)-conjugated secondary antibody from Santa Cruz Biotechnology (Frilabo, Porto, Portugal). For loading control, membranes were stripped and reprobed with a mouse monoclonal antiGAPDH antibody (6C5) from Santa Cruz Biotechnology (Frilabo, Porto, Portugal). The signal was detected with the ECL Amersham kit from GE Healthcare (VWR, Carnaxide, Portugal) and the Kodak GBX developer and fixer (Sigma–Aldrich, Sintra, Portugal). Band intensities were quantified using the Bio-Profil Bio-1D++ software (Vilber-Lourmat, Marne La Vallée, France). 2.2.8. Co-immunoprecipitation To perform the co-immunoprecipitation (co-IP), the Pierce Classic Magnetic IP and Co-IP Kit from Thermo scientific (Dagma, Carcavelos, Portugal) was used according to the manufacturer’s instructions. Briefly, cells were plated at a final density of 5  105 cells/plate and incubated for 24 h. Cells were then treated with 10 and 20 lM compound 3a, or DMSO only for 16 h. After cell lysis and protein lysate separation, 500 lg of total protein was incubated with 2 lL of mouse monoclonal anti-p53 (DO-1) or with mouse immunoglobulin G (IgG, negative control) from Santa Cruz Biotechnology (Frilabo, Porto, Portugal), overnight at 4 °C. Immunocomplexes were immunoprecipitated using magnetic beads. The Western blot analysis was performed as in Section 2.2.7. for whole cell lysate (input) and for immunoprecipitated proteins, detecting p53, MDM2 and GAPDH (loading control) proteins.

2.2.9. Flow cytometric analysis For the flow cytometric analysis, the FACSCalibur flow cytometer from BD Biosciences (Enzifarma, Porto, Portugal) and the CellQuest software from BD Biosciences (Enzifarma, Porto, Portugal) were used. Yeast and human tumor cell cycle phases were identified and quantified using ModFit LT software (Verity Software House Inc., Topsham, USA).

374

2.2.10. Statistical analysis Data were analyzed statistically using the GraphPad software. Differences between means were tested for significance using the Student’s t-test (⁄p < 0.05; ⁄⁄p < 0.01; ⁄⁄⁄p < 0.001).

381

2.3. Computational chemistry

385

The structures of the synthesized compounds 1a–c, 2a–b, 3a–b and 4a–b, Nutlin-3a and BZD were built with MOE. All the structures were energetically minimized using the MMFF94x force field with a RMS gradient of 0.1 and inserted in a test database. The crystallographic structure of MDM2 protein bound to the MI-63 analog inhibitor was downloaded from the Protein Data Bank (PDB code: 3LBL), imported in MOE and the co-crystallized waters were removed. After superposition of the A, C and E chains, it was verified that the two chains with a closer match were chains C and E. However, since the indole nitrogen of MI-63 analog is a hydrogen bond donor to LEU54, it was found that the one closer to the optimal distance of 2.8 Å, between donor and acceptor atoms, was the MI-63 analog interacting with chain C. Therefore, only the C chain was kept and the residues were protonated to match their state at physiological pH using the protonate3D module of MOE. The binding pocket was defined by using alpha spheres (sphere that contacts 4 atoms on its boundary and contains no internal atoms) within 5 Å of the MI-63 analog crystallographic pose. All the docking experiments were performed using a rigid protein docking protocol with a triangle matcher placement with a London dG scoring and a MMFF94x force field refinement. The best scored conformation per structure was selected and results were ranked through their binding affinities.

386

3. Results and discussion

410

3.1. Design and synthesis of a small library of phenylalaninol-derived oxazolopyrrolidone lactams to be evaluated as p53–MDM2 interaction inhibitors

411

Following our interest in the synthesis of novel p53–MDM2 interaction inhibitors (Ribeiro et al., 2014) we decided to synthesize a small library of phenylalaninol-derived oxazolopyrrolidone lactams. The oxazolo moiety with aromatic side chains attached would mimic the rigid heterocyclic scaffold present in most p53–MDM2 interaction inhibitors. The enantiopure phenylalaninol-derived oxazolopyrrolidone lactams were synthesized by cyclocondensation reaction between enantiopure phenylalaninol and different b-oxo-acids (Schemes 1 and 2) as previously described (Pereira et al., 2013). The reactions were performed in refluxing toluene for 16–48 h under a dean–stark apparatus and only one diastereoisomer was isolated. Compounds 1a–c were prepared in yields of 66–92% using S-phenylalaninol as chiral inducer and the appropriate b-oxo-benzoic acid derivatives (Scheme 1). Compounds 2a–b were prepared from reaction between S-phenylalaninol and the appropriate b-oxo-propanoic acids in yields of 73% and 72% for methyl (2a) and phenyl (2b) substitutions at C-7a, respectively (Scheme 1).

414

Q1 Please cite this article in press as: Soares, J., et al. Oxazoloisoindolinones with in vitro antitumor activity selectively activate a p53-pathway through potential inhibition of the p53–MDM2 interaction. Eur. J. Pharm. Sci. (2014), http://dx.doi.org/10.1016/j.ejps.2014.10.006

375 376 377 378 379 380

382 383 384

387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409

412 413

415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432

PHASCI 3120

No. of Pages 10, Model 5G

15 October 2014 J. Soares et al. / European Journal of Pharmaceutical Sciences xxx (2014) xxx–xxx

Scheme 1. Synthesis of compounds 1a–c and 2a–b. Reagents and conditions: (i) toluene, reflux, 66–92%.

Scheme 2. Synthesis of compounds 3a–b and 4a–b. Reagents and conditions: (i) toluene, reflux, 57–84%.

433 434 435 436 437 438 439 440 441

442 443 444 445 446 447 448 449 450

Bearing in mind the importance of the absolute stereochemical configuration of biologically active compounds, several cyclocondensation reactions using R-phenylalaninol as chiral inducer were also performed. Compounds 3a–b were prepared using benzoic acid derivatives as the b-oxo-acids in 74% yield (3a) for the enantiomer of compound 1b and 57% yield (3b) for the enantiomer of compound 1c (Scheme 2). Following the method above, compound 4a was prepared in 61% yield (enantiomer of compound 2a) and the compound 4b was obtained in 84% yield (Scheme 2). 3.2. Identification of oxazoloisoindolinones 1b and 3a as potential p53–MDM2 interaction inhibitors using a yeast-based assay In this work, a previously developed yeast-based assay (Leão et al., 2013a,c) was used for the analysis of the activity of the synthesized small library of phenylalaninol-derived oxazolopyrrolidone lactams as inhibitors of the p53–MDM2 interaction. The effectiveness of this yeast assay was recently demonstrated by the discovery of a new inhibitor of the p53–MDM2 interaction with a xanthone scaffold (Leão et al., 2013c). In this yeast assay, the

5

expression of human wt p53 induces growth inhibition and S-phase cell cycle arrest, which is abolished by co-expression with human MDM2 (Leão et al., 2013a,c). Inhibitors of the p53–MDM2 interaction, such as Nutlin-3a (positive control), reduce the impact of MDM2 on p53 activity, thus restoring the p53-induced yeast growth inhibition and cell cycle arrest (Fig. 1A and B). Notably, these small molecules do not interfere with the activity of p53 when expressed alone (Leão et al., 2013a,c). Using this approach, the effect of 1–50 lM of compounds was evaluated using, as endpoint, the MDM2-dependent reduction of p53-induced yeast growth inhibition. We started by testing five oxazolopyrrolidone lactams derived from S-phenylalaninol. From the concentration–response curves obtained for the tested compounds, the concentration of 10 lM was selected as the lowest concentration at which a higher reduction of the MDM2 effect was obtained (see Fig. 1A for compounds 1b and 3a), without interfering with the growth of control yeast (empty vectors). Among the tested compounds, oxazoloisoindolinone 1b significantly reverted the inhibitory effect of MDM2 on p53-induced yeast growth inhibition at a wide range of concentrations tested (Fig. 1A), exhibiting a similar effect to Nutlin-3a (Table 1). Accordingly, like Nutlin-3a, compound 1b significantly reduced the inhibitory effect of MDM2 on p53-induced S-phase cell cycle arrest (Fig. 1B). Nevertheless, compound 1b did not interfere with the activity of p53 when expressed alone and with the growth of control yeast (vector) (Fig. 1C). With the results obtained in yeast, we could conclude that a phenyl group is preferred to a methyl group at position R1 (compounds 1b and 2b versus compounds 1a and 2a) (Table 1). Also, compounds 1 containing an isoindolinone moiety are more active than compounds 2 with a pyrrolidone moiety (Table 1). Due to the importance of chirality on the biological activity, it was pivotal to assess the potential value of the enantiomers of the most active compounds. Four oxazolopyrrolidones derived from R-phenylalaninol were chosen to be analyzed on the p53– MDM2 interaction. In general, all compounds were more active than the S-phenylalaninol counterparts. In addition, as observed with compound 1b, the compound 3a significantly reverted the inhibitory effect of MDM2 on p53-induced yeast growth inhibition at a wide range of concentrations tested (Fig. 1A), exhibiting a similar effect to Nutlin-3a (Table 2). Accordingly, compound 3a significantly reduced the inhibitory effect of MDM2 on p53-induced Sphase cell cycle arrest (Fig. 1B). However, compound 3a did not interfere with the activity of p53 when expressed alone and with the growth of control yeast (vector) (Fig. 1C). In order to check the selectivity of oxazoloisoindolinone 1b and 3a for MDM2, the effect of these two compounds on the other p53 binding partner, MDMX, was also checked, using a previously developed yeast assay to search for inhibitors of the p53–MDMX interaction (Leão et al., 2013b). Contrary to the well-known inhibitor of the p53–MDMX interaction (SJ-172550, positive control) (Leão et al., 2013b), oxazoloisoindolinone 1b and 3a did not behaved as potential inhibitors of the p53–MDMX interaction, since they did not interfere with the negative effect of MDMX on p53 in yeast (data not shown). Together, the results obtained in yeast identified oxazoloisoindolinones 1b and 3a as potential inhibitors of the p53–MDM2 interaction.

451

3.3. Oxazoloisoindolinone 3a in vitro antitumor activity involves selective activation of a p53–pathway and inhibition of the p53– MDM2 interaction

509

The tumor cell growth inhibitory potential of oxazoloisoindolinones 1b and 3a and the contribution of the p53-pathway to their activities were thereafter ascertained using human colon

512

Q1 Please cite this article in press as: Soares, J., et al. Oxazoloisoindolinones with in vitro antitumor activity selectively activate a p53-pathway through potential inhibition of the p53–MDM2 interaction. Eur. J. Pharm. Sci. (2014), http://dx.doi.org/10.1016/j.ejps.2014.10.006

452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508

510 511

513 514

PHASCI 3120

No. of Pages 10, Model 5G

15 October 2014 6

J. Soares et al. / European Journal of Pharmaceutical Sciences xxx (2014) xxx–xxx

Fig. 1. Identification of oxazoloisoindolinones 1b and 3a as potential inhibitors of the p53–MDM2 interaction using the yeast-based assay. (A) Percentage of reversion of the MDM2 negative effect on p53-induced growth inhibition caused by compounds 1b and 3a and Nutlin-3a (positive control); the growth of yeast cells co-expressing p53 and MDM2 treated with 1–50 lM of compounds or DMSO only, for 42 h, was analysed by CFU counts; results are plotted as relative to the growth inhibition achieved with yeast cells expressing p53 alone treated with DMSO only (set to 100%); presented data are mean ± S.E.M. of five independent experiments. (B) Effect of compounds 1b and 3a on the inhibition of p53-induced yeast cell cycle arrest by MDM2. Yeast cells were incubated in the presence of 10 lM of compounds 1b and 3a, 10 lM of Nutlin-3a or DMSO only, for 42 h. Cell cycle phases were quantified by flow cytometry; presented data are mean ± S.E.M. of two independent experiments. (C) Effect of compounds 1b and 3a on the growth of yeast cells expressing p53 alone and control yeast (vector); growth curves were obtained by CFU counts for yeast cells expressing p53 alone and control yeast treated with 1–50 lM of compounds 1b and 3a or DMSO only, for 42 h; presented data are mean ± S.E.M. of five independent experiments. In (A) and (B), values are significantly different from DMSO only (⁄p < 0.05, ⁄⁄p < 0.001).

Table 1 Effect of compounds 1a–c and 2a–b on the reversion of p53-induced yeast growth inhibition by MDM2.

Compounds

R1

R2

Reversion of MDM2 effect (%)a

1a 1b 1c 2a 2b

Me Ph H Me Ph

H H OMe – –

53.6 ± 3.0 82.5 ± 6.4 69.9 ± 3.5 38.7 ± 3.2 47.6 ± 6.1

DMSO Nutlin-3a

– –

– –

2.3 ± 5.6 73.2 ± 3.0

a Yeast cells co-expressing p53 and MDM2 were incubated in selective medium in the presence of 10 lM compound or DMSO only, for 42 h. Nutlin-3a was used as positive control. The growth inhibition obtained with yeast cells expressing only p53 treated with DMSO was considered as 100%. Data are mean ± S.E.M. of five independent experiments.

515 516 517 518 519 520

adenocarcinoma HCT116 cell lines with wt p53 (HCT116 p53+/+) and its isogenic derivative, in which p53 has been knocked out (HCT116 p53 / ). The GI50 (concentration that causes 50% growth inhibition) values obtained for compound 1b in p53+/+ (8.8 lM) and p53 / (11.4 lM) HCT116 cells indicated a potent growth inhibitory effect against these tumor cells (Fig. 2). In spite of this,

Table 2 Effect of compounds 3a–b and 4a–b on the reversion of p53-induced yeast growth inhibition by MDM2.

Compounds

R1

R2

Reversion of MDM2 effect (%)a

3a 3b 4a 4b

Ph H Me Ph

H OMe – –

88.1 ± 1.9 66.8 ± 5.0 66.5 ± 4.7 75.1 ± 4.0

a Yeast cells co-expressing p53 and MDM2 were incubated in selective medium in the presence of 10 lM compound or DMSO only, for 42 h. Nutlin-3a was used as positive control. The growth inhibition obtained with yeast cells expressing only p53 treated with DMSO was considered as 100%. Data are mean ± S.E.M. of five independent experiments.

contrary to Nutlin-3a, no differences in the activity of compound 1b were observed between p53+/+ and p53 / HCT116 cells. Together with the results obtained in yeast, we could conclude that compound 1b was a nonselective activator of the p53-pathway in human tumor cells. Interestingly, similar results were recently reported for an optimized isoindolinone compound obtained from a serial of structural modifications, which despite its high potency, revealed to be nonselective for the p53-pathway (Hardcastle et al., 2011).

Q1 Please cite this article in press as: Soares, J., et al. Oxazoloisoindolinones with in vitro antitumor activity selectively activate a p53-pathway through potential inhibition of the p53–MDM2 interaction. Eur. J. Pharm. Sci. (2014), http://dx.doi.org/10.1016/j.ejps.2014.10.006

521 522 523 524 525 526 527 528 529

PHASCI 3120

No. of Pages 10, Model 5G

15 October 2014 J. Soares et al. / European Journal of Pharmaceutical Sciences xxx (2014) xxx–xxx

Fig. 2. Oxazoloisoindolinone 3a exhibits in vitro antitumor activity through selective activation of a p53-dependent pathway in human tumor cell lines. The GI50 concentration was determined for compounds 1b, 3a and Nutlin-3a, in human colon adenocarcinoma HCT116 cell lines with (p53+/+) and without (p53 / ) p53, after 48 h incubation, using the sulforhodamine B assay. Presented data are mean ± S.E.M. of four independent experiments; values are significantly different from HCT116 p53+/+ cells (⁄⁄p < 0.01).

530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548

Although a similar GI50 value to that of the compound 1b was obtained with compound 3a (9.7 lM) in HCT116 p53+/+, a significant increase in the selectivity of compound 3a to the p53-pathway was observed (Fig. 2). In fact, as observed with Nutlin-3a, a significant reduction of the potency of compound 3a was observed in HCT116 p53 / cell lines. The molecular mechanism of action of oxazoloisoindolinone 3a as a potential inhibitor of the p53–MDM2 interaction was thereafter investigated in HCT116 tumor cell lines. In HCT116 p53+/+ cells, the growth inhibitory effect of compound 3a, at GI50 (9.7 lM) and twice the GI50 (2xGI50) concentration, was associated to the induction of G0/G1-phase cell cycle arrest (Fig. 3A). Accordingly, 9.7 lM of compound 3a up-regulated the p21 expression levels (a key regulator of cell cycle progression and a p53 target gene) in p53+/+, but not in p53 / , HCT116 cells (Fig. 3B). Additionally, it was observed that the growth inhibitory effect of compound 3a in HCT116 p53+/+ cells was also associated to the induction of late apoptosis both at GI50 and 2xGI50 concentration (Fig. 4A). The activation of a p53-dependent apoptotic pathway

7

by 9.7 lM of compound 3a was further supported by the occurrence of PARP cleavage (a result of caspase-3 activation) (Fig. 4B) and by the up-regulation of p53 downstream pro-apoptotic targets of the Bcl-2 family, BAX and PUMA (Fig. 5A) only in p53-expressing HCT116 cells. Additionally, 9.7 lM of compound 3a also up-regulated the expression levels of MDM2, other p53 target gene, and led to an increase of the p53 baseline levels in p53+/+, but not in p53 / , HCT116 cells (Fig. 5A). This stabilization of p53 protein levels indicates an inhibition by compound 3a of the p53 degradation caused by its interaction with MDM2 in these cells. Together, these results indicated that oxazoloisoindolinone 3a reduced the p53 inhibition by MDM2, promoting its transcriptional activity and subsequently increasing the expression levels of p53 target genes. Moreover, the ability of oxazoloisoindolinone 3a to block the intracellular p53–MDM2 interaction in HCT116 p53+/+ cells was checked by co-IP. The results obtained confirmed that 10 and 20 lM compound 3a blocked the p53–MDM2 interaction (Fig. 5B). In fact, these experiments showed that little or no MDM2 protein co-immunoprecipitated with p53 in HCT116 p53+/+ cells treated with oxazoloisoindolinone 3a.

549

3.4. Predicted binding model supports the oxazoloisoindolinone 3a binding to MDM2

570

Molecular docking studies with the tested phenylalaninolderived oxazolopyrrolidone lactams were performed on the MDM2 hydrophobic cleft using the Molecular Operating Environment (MOE) software and the crystallographic structure with PDB code 3LBL (www.pdb.org) (Berman et al., 2000) allowing to rank the compounds by binding affinities summarized in Table 3. Only the best scoring conformation per molecule was considered in the discussion. The co-crystallized ligand MI-63 analog was re-docked as a way to validate the molecular docking method. The root mean square deviation (RMSD) between the predicted and co-crystallized pose of MI-63 analog was 1.11 Å (predicted binding energy: 29.37 kcal/mol) which indicates that the software is able to accurately reproduce the binding pose of the co-crystallized ligand. A second validation of the docking method was performed by confirming that the interactions observed for the docked Nutlin-3a and BZD structures were in agreement with the interactions found for these two compounds in other crystallographic structures (PDB code: 4J3E, for Nutlin-3a; PDB code: 1T4E, for BZD). The values obtained for inhibitors which are known to establish the well characterized interactions with the MDM2 protein were found to

572

Fig. 3. Oxazoloisoindolinone 3a induces cell cycle arrest in human tumor cells. (A) Effect of compound 3a on cell cycle progression of HCT116 p53+/+ cells was determined after 48 h treatment with 9.7 lM of compound 3a and 3.2 lM of Nutlin-3a (GI50 values), or with 2  GI50. Cell cycle phases were analyzed by flow cytometry using propidium iodide (PI) and quantified using ModFit LT software. Presented data are mean ± S.E.M. of two independent experiments; values are significantly different from DMSO (⁄p < 0.05; ⁄⁄p < 0.01). (B) Compound 3a increases the p21 protein levels in p53+/+, but not in p53 / , HCT116 cells. Western blot analysis was performed after 4 h treatment with 9.7 lM of compound 3a or DMSO only. The quantification of band intensity was performed using GAPDH as loading control.

Q1 Please cite this article in press as: Soares, J., et al. Oxazoloisoindolinones with in vitro antitumor activity selectively activate a p53-pathway through potential inhibition of the p53–MDM2 interaction. Eur. J. Pharm. Sci. (2014), http://dx.doi.org/10.1016/j.ejps.2014.10.006

550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569

571

573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592

PHASCI 3120

No. of Pages 10, Model 5G

15 October 2014 8

J. Soares et al. / European Journal of Pharmaceutical Sciences xxx (2014) xxx–xxx

Fig. 4. Oxazoloisoindolinone 3a induced early and late apoptosis in HCT116 p53+/+ cells. (A) Apoptosis was determined after 48 h treatment with 9.7 lM of compound 3a and 3.2 lM of Nutlin-3a (GI50 values) or with 2  GI50 and was analyzed by flow cytometry using FITC-Annexin V and PI; quantification of cells in apoptosis are mean ± S.E.M. of two independent experiments; values are significantly different from DMSO (⁄p < 0.05; ⁄⁄p < 0.01). (B) Compound 3a leads to PARP cleavage in p53+/+, but not in p53 / , HCT116 cells. Western blot analysis was performed after 24 h treatment with 9.7 lM of compound 3a or DMSO only. The quantification of band intensity was performed using GAPDH as loading control.

Fig. 5. Oxazoloisoindolinone 3a increases the expression levels of p53 and of proteins encoded by p53 target genes in p53+/+, but not in p53 / , HCT116 cells and disrupted the intracellular p53–MDM2 interaction. (A) Western blot analysis was performed after 8 h (for p53) and 24 h (for Bax, MDM2 and PUMA) treatments with 9.7 lM of compound 3a or DMSO only. The quantification of band intensity was performed using GAPDH as loading control. (B) HCT116 p53+/+ cells were treated with 10 and 20 lM oxazoloisoindolinone 3a or DMSO only for 16 h, followed by IP with p53 antibody or mouse immunoglobulin G (IgG) antibody and immunoblot with MDM2 and p53 antibodies; whole cell lysate (input). GAPDH was used as loading control.

Table 3 Binding affinities found by molecular docking for the tested compounds. Compounds

DG (kcal/mol)

3a 1b 4a 1c 4b 2b 1a 2a 3b

21.61 20.80 18.34 18.11 16.89 16.80 16.77 15.74 14.82

Nutlin-3a 2-Cl-BZD

26.50 21.91

be between 26.50 (Nutlin-3a) and 21.91 (BZD) kcal/mol. Among the tested phenylalaninol-derived oxazolopyrrolidone lactams, oxazoloisoindolinones 3a and 1b presented values of 21.61 and 20.80 kcal/mol respectively, thus having the best binding affinities within this new group of p53–MDM2 interaction inhibitors (Table 3). It is important to stress that these compounds were also the ones having the highest activities in the yeast screening assay and that a good correlation was found between the binding affinities estimated by docking for this class of compounds and the percentage of inhibition. We have visually inspected the binding interactions established between compound 3a and MDM2 protein, and the conformation with the highest score is displayed in Fig. 6. It was found that for the most active compound, 3a, the binding interactions are in agreement with one of the possible

Q1 Please cite this article in press as: Soares, J., et al. Oxazoloisoindolinones with in vitro antitumor activity selectively activate a p53-pathway through potential inhibition of the p53–MDM2 interaction. Eur. J. Pharm. Sci. (2014), http://dx.doi.org/10.1016/j.ejps.2014.10.006

593 594 595 596 597 598 599 600 601 602 603 604 605 606

PHASCI 3120

No. of Pages 10, Model 5G

15 October 2014 J. Soares et al. / European Journal of Pharmaceutical Sciences xxx (2014) xxx–xxx

Fig. 6. Docking pose of oxazoloisoindolinone 3a within the MDM2 hydrophobic cleft limits depicted with a surface (in green, hydrophobic; in pink, hydrophilic areas). Receptor carbon residues are displayed in yellow and LEU54 and ILE61 are highlighted. Part of the p53 backbone (helix in red, turn in blue, loop in white) and residues PHE19, TRP23 and LEU26 are represented with carbons in light blue. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643

binding modes already described for isoindolinones derivatives, found through NMR studies (Riedinger et al., 2008). The aromatic ring of the isoindolinone core establishes hydrophobic contacts with residue ILE61. The phenyl group at position C9b is projected towards the deep hydrophobic pocket of MDM2 and the methylene-phenyl moiety establishes hydrogen-p interactions with residue LEU54. The 1YCR structure containing p53 interacting with MDM2 was aligned and superposed with the 3LBL structure using the MDM2 residues defining the pocket. The alignment and the positioning of residues responsible for the most important interactions are depicted in Fig. 6. The picture makes clear the mimetic nature of compound 3a where the location occupied by PHE19 is now occupied by the aromatic ring of isoindolinone moiety, the TRP23 by the phenyl group at position 9b and LEU26 by the methylphenyl group. In protein–protein interactions, there are some concerns about possible conformational changes produced on the binding site upon binding to proteins or ligands that will impose the usage of other research techniques besides docking. The binding site defined by the area where all organic molecules mimicking p53 interact is not very deep like a pocket, but resembles a cleft/groove and is hydrophobic in nature. However, the organic molecules all have a ring or fused rings located deeply in the pocket where PHE19 of p53 interact. In opposition to the other two pockets (defined by TRP23 and LEU26 of p53), this one goes deep in the protein structure and seems to work as an anchor point guiding the positioning of the binding molecules. For instance, superposing structures 1YCR (p53 co-crystallized) with 1T4E (benzodiazepine co-crystallized) shows a relatively low RMSD for the alpha carbons (0.9 Å). The amino acids of the three pockets defined by p53 have very similar rotameric states for the residues, defining the three referred pockets except for Tyr67, HIS73 and MET62. The latter is the only residue in a different rotameric state not belonging to a coil region found between TYR67 and HIS73. MET62 is the only residue in a different rotameric state not belonging to the coil region. All these residues have no specific

9

interactions with the co-crystallized molecules except a generic hydrophobic match. Moreover, the synthesized molecules reported in the manuscript also have no direct interactions with this residue and the closest atoms are characteristically at a distance of at least 4 Å. Therefore, the possibility of using molecular docking to predict the binding modes of the newly synthesized molecules seems adequate. These observations provide insights about the binding mode of the most active hit in this study. Therefore, it is expected to allow a more guided structure optimization of compound 3a. Based on this, efforts are currently underway for the development of more potent p53–MDM2 interaction inhibitors.

644

4. Conclusion

656

In this work, nine phenylalaninol-derived oxazolopyrrolidone lactams were synthesized to probe their capacity of inhibiting the p53–MDM2 interaction. The screening of the compounds using a yeast cell model led us to the identification of two potential inhibitors of the p53–MDM2 interaction. In conformity to what was obtained in yeast, oxazoloisoindolinone 3a mimicked the activity of known small molecule inhibitors of the p53–MDM2 interaction in human tumor cells, leading to the successful activation of p53 and downstream cell signaling. In fact, in tumor cells, the oxazoloisoindolinone 3a exhibited the most relevant hallmarks of a typical inhibitor of the p53–MDM2 interaction. Particularly, it decreased the proliferation of wt p53-carrying tumor cells, presenting a more reduced activity in tumor cells without p53. Additionally, in wt p53-carrying tumor cells, it induced cell cycle arrest and apoptosis, increased the p53 protein levels, and up-regulated the p53 transcriptional activity with an increase of the expression levels of p53 target genes (MDM2, p21, BAX and PUMA). Most importantly, co-IP assays reinforced the ability of oxazoloisoindolinone 3a to inhibit the p53–MDM2 interaction in tumor cells. Finally, molecular docking studies supported the binding of oxazoloisoindolinone 3a to MDM2. Altogether, this work reveals a new class of potent and selective small molecule activators of the p53-pathway with promising antitumor activity. Additionally, it may pave the way to the development of a new class of p53–MDM2 interaction inhibitors In fact, oxazoloisoindolinone 3a represents a lead compound for the structure-based design of analogs with improved pharmacological properties than those currently available.

657

Acknowledgments

685

This work received the financial support from the European Union (FEDER funds through COMPETE) and National Funds (FCT, Fundação para a Ciência e Tecnologia) through REQUIMTE (PestC/EQB/LA0006/2013), iMed.ULisboa (Pest-OE/SAU/UI4013/2014), and the research project PTDC/QUI-QUI/111664/2009, and FCOMP01-0124-FEDER-015752 (PTDC/SAU-FAR/110848/2009), by U. Porto/ Santander Totta, and in part by the Italian Association for Cancer Research, AIRC (IG #9086). It was also supported by FCT fellowships to J. Soares (SFRH/BD/78971/2011), N.A.L. Pereira (PTDC/QUI-QUI/ 111664/2009), A. Monteiro (PTDC/SAU-FAR/110848/2009), M. Leão (SFRH/BD/64184/2009) and C. Bessa (SFRH/BD/87109/2012).

686

Appendix A. Supplementary material

697

Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.ejps.2014.10.006.

698

Q1 Please cite this article in press as: Soares, J., et al. Oxazoloisoindolinones with in vitro antitumor activity selectively activate a p53-pathway through potential inhibition of the p53–MDM2 interaction. Eur. J. Pharm. Sci. (2014), http://dx.doi.org/10.1016/j.ejps.2014.10.006

645 646 647 648 649 650 651 652 653 654 655

658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684

687 688 689 690 691 692 693 694 695 696

699

PHASCI 3120

No. of Pages 10, Model 5G

15 October 2014 10

J. Soares et al. / European Journal of Pharmaceutical Sciences xxx (2014) xxx–xxx

700

References

701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735

Allin, S.M., James, S.L., Martin, W.P., Smith, T.A.D., Elsegoog, M.R.J., 2001. Stereoselective synthesis of the pyrroloisoquinoline ring system. J. Chem. Soc. Perkin Trans. 22, 3029–3036. Berman, H.M., Westbrook, J., Feng, Z., Gilliland, G., Bhat, T.N., Weissig, H., Shindyalov, I.N., Bourne, P.E., 2000. The Protein Data Bank. Nucleic Acids Res. 28, 235–242. Hardcastle, I.R., Ahmed, S.U., Atkins, H., Calvert, A.H., Curtin, N.J., Farnie, G., Golding, B.T., Griffin, R.J., Guyenne, S., Hutton, C., Källblad, P., Kemp, S.J., Kitching, M.S., Newell, D.R., Norbedo, S., Northen, J.S., Reid, R.J., Saravanan, K., Willems, H.M., Lunec, J., 2005. Isoindolinone-based inhibitors of the MDM2–p53 protein– protein interaction. Bioorg. Med. Chem. Lett. 15, 1515–1520. Hardcastle, I.R., Liu, J., Valeur, E., Watson, A., Ahmed, S.U., Blackburn, T.J., Bennaceur, K., Clegg, W., Drummond, C., Endicott, J.A., Golding, B.T., Griffin, R.J., Gruber, J., Haggerty, K., Harrington, R.W., Hutton, C., Kemp, S., Lu, X., McDonnell, J.M., Newell, D.R., Noble, M.E., Payne, S.L., Revill, C.H., Riedinger, C., Xu, Q., Lunec, J., 2011. Isoindolinone inhibitors of the murine double minute 2 (MDM2)–p53 protein–protein interaction: structure–activity studies leading to improved potency. J. Med. Chem. 54, 1233–1243. Kamal, A., Mohammed, A.A., Shaik, T.B., 2012. P53–Mdm2 inhibitors: patent review (2009–2010). Expert Opin. Ther. Pat. 22, 95–105. Lauria, A., Tutone, M., Ippolito, M., Pantano, L., Almerico, A.M., 2011. Molecular modeling approaches in the discovery of new drugs for anti-cancer therapy: the investigation of p53–MDM2 interaction and its inhibition by small molecules. Curr. Med. Chem. 17, 3142–3154. Leão, M., Gomes, S., Pedraza-Chaverri, J., Machado, N., Sousa, E., Pinto, M., Inga, A., Pereira, C., Saraiva, L., 2013a. F-mangostin and gambogic acid as potential inhibitors of the p53–MDM2 interaction revealed by a yeast approach. J. Nat. Prod. 76, 774–778. Leão, M., Gomes, S., Soares, J., Bessa, C., Maciel, C., Ciribilli, Y., Pereira, C., Inga, A., Saraiva, L., 2013b. Novel simplified yeast-based assays of regulators of p53– MDMX interaction and p53 transcriptional activity. FEBS J. 280, 6498–6507. Leão, M., Pereira, C., Bisio, A., Ciribilli, Y., Paiva, A.M., Machado, N., Palmeira, A., Fernandes, M.X., Sousa, E., Pinto, M., Inga, A., Saraiva, L., 2013c. Discovery of a new small-molecule inhibitor of p53–MDM2 interaction using a yeast-based approach. Biochem. Pharmacol. 85, 1234–1245.

Pereira, N.A.L., Sureda, F.X., Turch, M., Amat, M., Bosch, J., Santos, M.M.M., 2013. Synthesis of phenylalaninol-derived oxazolopyrrolidone lactams and evaluation as NMDA receptor antagonists. Monatsh. Chem. 144, 473–477. Ray-Coquard, I., Blay, J.Y., Italiano, A., Le Cesne, A., Penel, N., Zhi, J., Heil, F., Rueger, R., Graves, B., Ding, M., Geho, D., Middleton, S.A., Vassilev, L.T., Nichols, G.L., Bui, B.N., 2012. Effect of the MDM2 antagonist RG7112 on the P53 pathway in patients with MDM2-amplified, well-differentiated or dedifferentiated liposarcoma: an exploratory proof-of-mechanism study. Lancet Oncol. 13, 1133–1140. Rew, Y., Sun, D., Gonzalez-Lopez De Turiso, F., Bartberger, M.D., Beck, H.P., Canon, J., Chen, A., Chow, D., Deignan, J., Fox, B.M., Gustin, D., Huang, X., Jiang, M., Jiao, X., Jin, J., Kayser, F., Kopecky, D.J., Li, Y., Lo, M.C., Long, A.M., Michelsen, K., Oliner, J.D., Osgood, T., Ragains, M., Saiki, A.Y., Schneider, S., Toteva, M., Yakowec, P., Yan, X., Ye, Q., Yu, D., Zhao, X., Zhou, J., Medina, J.C., Olson, S.H., 2012. Structurebased design of novel inhibitors of the MDM2–p53 interaction. J. Med. Chem. 55, 4936–4954. Ribeiro, C.J.A., Amaral, J.D., Rodrigues, C.M.P., Moreira, R., Santos, M.M.M., 2014. Synthesis and evaluation of spiroisoxazoline oxindoles as anticancer agents. Bioorg. Med. Chem. 22, 577–584. Riedinger, C., Endicott, J.A., Kemp, S.J., Smyth, L.A., Watson, A., Valeur, E., Golding, B.T., Griffin, R.J., Hardcastle, I.R., Noble, M.E., McDonnell, J.M., 2008. Analysis of chemical shift changes reveals the binding modes of isoindolinone inhibitors of the MDM2–p53 interaction. J. Am. Chem. Soc. 130, 16038–16044. Vassilev, L.T., Vu, B.T., Graves, B., Carvajal, D., Podlaski, F., Filipovic, Z., Kong, N., Kammlott, U., Lukacs, C., Klein, C., Fotouhi, N., Liu, E.A., 2004. In vivo activation of the p53 pathway by small-molecule antagonists of MDM2. Science 6, 844–848. Wang, W., Hu, Y., 2012. Small molecule agents targeting the p53–MDM2 pathway for cancer therapy. Med. Res. Rev. 32, 1159–1196. Wang, S., Zhao, Y., Bernard, D., Aguilar, A., Kumar, S., 2012. Targeting the MDM2– p53 protein–protein interaction for new cancer therapeutics. Top. Med. Chem. 8, 57–80. Zhao, Y., Bernard, D., Wang, S., 2013a. Small Molecule inhibitors of MDM2–p53 and MDMX–p53 interaction as new cancer therapeutics. BioDiscovery 8, 4. Zhao, Y., Yu, S., Sun, W., Liu, L., Lu, J., McEachern, D., Shargary, S., Bernard, D., Li, X., Zhao, T., Zou, P., Sun, D., Wang, S., 2013b. A potent small-molecule inhibitor of the MDM2–p53 interaction (MI-888) achieved complete and durable tumor regression in mice. J. Med. Chem. 56, 5553–5561.

736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773

Q1 Please cite this article in press as: Soares, J., et al. Oxazoloisoindolinones with in vitro antitumor activity selectively activate a p53-pathway through potential inhibition of the p53–MDM2 interaction. Eur. J. Pharm. Sci. (2014), http://dx.doi.org/10.1016/j.ejps.2014.10.006