CBI 7321

No. of Pages 17, Model 5G

31 March 2015 Chemico-Biological Interactions xxx (2015) xxx–xxx 1

Contents lists available at ScienceDirect

Chemico-Biological Interactions journal homepage: www.elsevier.com/locate/chembioint

6 7 3 4 5 8 9 10 12 11 13 1 3 5 2 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31

Mechanisms of DNA methyltransferase–inhibitor interactions: Procyanidin B2 shows new promise for therapeutic intervention of cancer Arunima Shilpi, Sabnam Parbin, Dipta Sengupta, Swayamsiddha Kar, Moonmoon Deb, Sandip Kumar Rath, Nibedita Pradhan, Madhumita Rakshit, Samir Kumar Patra ⇑ Epigenetics and Cancer Research Laboratory, Biochemistry and Molecular Biology Group, Department of Life Science, National Institute of Technology, Rourkela, Odisha 769008, India

a r t i c l e

i n f o

Article history: Received 10 August 2014 Received in revised form 17 March 2015 Accepted 22 March 2015 Available online xxxx Keywords: DNA methyltransferases Cancer Procyanidin B2 Epigallocathechin-3-gallate Molecular docking Simulation MM-GBSA MM-PBSA DNMT activity assay

a b s t r a c t DNA methyltransferases (DNMTs) is a key epigenetic enzyme for pharmacological manipulation and is employed in cancer reprogramming. During past few years multiple strategies have been implemented to excavate epigenetic compounds targeting DNMTs. In light of the emerging concept of chemoinformatics, molecular docking and simulation studies have been employed to accelerate the development of DNMT inhibitors. Among the DNMT inhibitors known till date, epigallocathechin-3-gallate (EGCG) was identified to be effective in reducing DNMT activity. However, the broad spectrum of EGCG to other diseases and variable target enzymes offers some limitations. In view of this, 32 EGCG analogues were screened at S-Adnosyl-L-homocysteine (SAH) binding pocket of DNMTs and procyanidin B2–3, 30 -di-Ogallate (procyanidin B2) was obtained as potent inhibitor having medicinally relevant chemical space. Further, in vitro analysis demonstrates the efficiency of procyanidin B2 in attenuating DNMT activity at IC50 of 6.88 ± 0.647 lM and subsequently enhancing the expression of DNMT target genes, E-cadherin, Maspin and BRCA1. Moreover, the toxic property of procyanidin B2 towards triple negative breast cancer cells to normal cells offers platform for pre-clinical trial and an insight to the treatment of cancer. Ó 2015 Elsevier Ireland Ltd. All rights reserved.

33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49

50 51

1. Introduction

52

In vertebrates, the most significant form of DNA methylation is associated with 5-carbon position of cytosine residues at CpG dinucleotide sequences clustered as CpG islands, located mostly in the promoter region of genes. Increased levels of cytosine methylation of promoter regions are strongly associated with transcription repression [1,2]. The transcriptional activation or silencing of genes are associated with several biological processes, such as embryonic development [3,4], stem cell differentiation [5], genomic imprinting [6], transposable elements silencing [7],

53 54 55 56 57 58 59 60

Abbreviations: DNMT, DNA methyltransferases; EGCG, epigallocathechin-3gallate; SAH, S-Adnosyl-L-homocysteine; ChEBI, chemical entities of biological interest; PDB, protein data bank; UniProt, the Universal Protein knowledgebase; CHARMm, chemistry at HARvard molecular mechanics; PLP, piecewise linear potential; PMF, potentials of mean force; LC50, lethal constant at 50% concentration; IC50, inhibition constant at which the enzyme exhibit 50% activity; MD, molecular dynamics; RMSD, root mean square deviation; RMSF, root mean square fluctuation. ⇑ Corresponding author. Tel.: +91 6612462683; fax: +91 6612462681. E-mail addresses: [email protected], [email protected] (S.K. Patra).

ageing, inflammation and cancer [8,9]. Aberrant gene expression, genome wide hypomethylation and regional hypermethylation is frequently observed in human cancers [10]. The significance of CpG hypermethylation in cancer is characterised by frequency of its occurrence in various genes of multiple cancers [11]. The reversal of these hypermethylation by small-molecule inhibitors may provide novel cancer therapeutic strategies [12]. Cellular DNA methylation is established and maintained by complex interplay of family of dedicated enzymes called DNA methyltransferases (DNMTs) [13]. In mammals, three DNMTs, namely, DNMT1, DNMT3A and DNMT3B are most significant with structural and functional distinctions. DNMT1 is the most abundant methyltransferases in somatic cell, responsible for restoring post replicative hemi-methylation to full DNA methylation, thus, referred to as maintenance methyltransferases; while DNMT3a and DNMT3b are responsible for de novo methylation that occurs in genome following embryo implantation [14,15]. These enzymes catalyse the transfer of methyl group from the co-factor S-Adenosyl methionine (SAM/AdoMet). The most significant feature of this catalytic process is nucleophilic attack by thiol group

http://dx.doi.org/10.1016/j.cbi.2015.03.022 0009-2797/Ó 2015 Elsevier Ireland Ltd. All rights reserved.

Please cite this article in press as: A. Shilpi et al., Mechanisms of DNA methyltransferase–inhibitor interactions: Procyanidin B2 shows new promise for therapeutic intervention of cancer, Chemico-Biological Interactions (2015), http://dx.doi.org/10.1016/j.cbi.2015.03.022

61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80

CBI 7321

No. of Pages 17, Model 5G

31 March 2015 2 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145

A. Shilpi et al. / Chemico-Biological Interactions xxx (2015) xxx–xxx

of cysteine residue on C-6 carbon of cytosine nucleotide forming DNA-enzyme covalent adduct. The covalent bond formation in turn activates the C-5 atom towards electrophilic attack which leads to the addition of methyl group to 5-carbon by elimination of proton and formation of S-Adnosyl-L-homocysteine (SAH). The glutamic acid residue present in the active site (motif VI) is significant, as it stabilises DNA–protein complex. Moreover, high energy carbanion is stabilised by resonance mechanism where arginine residue plays a significant role [16,17]. Thus, presence of Cys, Glu and Arg residues in the active site promotes active DNA methylation. The active site of DNMTs constitute six conserved motifs, namely I, IV, VI, VIII, IX and X [18]. Along with regional hypermethylation and overall hypomethylation of the genome, in many cancers, it has been reported that the expression and activity of DNMTs is very high [19]. This gives a clue that DNMTs may have oncogenic potential, apart from its DNA methylation activity, and thus DNMTs emerged as a potential anticancer drug development target. For therapeutic intervention targeting enzymes inhibitors are essential which bind to the catalytic domain of the enzymes (in this case, DNMTs) and inhibit its activity [20]. Currently, the available inhibitors of DNMTs are classified into two broad groups, known as nucleoside and non-nucleoside analogues. The archetypal nucleoside inhibitors are derivatives of the cytidine nucleoside and they inhibit DNMT activity only after getting incorporated into newly synthesized DNA strands and trapping the enzyme by forming DNA–protein adduct [20,21]. Chemically these cytosine analogues constitute 5Azacytidine (Vidaza) [22], Decitabine (Dacogen) [23], Zebularine and 5-Fluoro-2 deoxycytidine [24]. However, the non-nucleoside inhibitors directly block DNMT activity by binding to its active site pockets or cofactor binding pockets. Among the known non-nucleoside inhibitors, RG108 and NSC14778 are synthetic analogues while the major group constitute the phytochemicals. Some of these phytochemicals, like, epigallocathechin-3-gallate (EGCG), tea polyphenols [25]; mahanine, a carbazole alkaloid [26]; and curcumin, a component of the Indian spice turmeric [27] have been reported to possess anticancer properties. Many others are effective against non-cancerous diseases. Some of them are hydralazine, as antihypertensive drug [28]; procaine, as local anaesthetic [29] and procainamide, as antiarrhythmic drug [30]. SAH, the end product of DNA methylation reaction, and its analogues apparently have been reported as selective inhibitors towards inhibition of DNA methylation as they bind to the co-factor binding pocket of DNMTs [31,32]. In this investigation, we report for the first time a detail view of the elementary interactions between non-nucleoside inhibitors and DNMTs in the active site terminal and thus find out the best inhibitors. To explore the interactions between the enlisted nonnucleoside inhibitors (Fig. 1) and DNMTs, we performed docking and simulation studies. Application of docking using various algorithms affirms the binding pattern and profile of a ligand. The relative binding site for all the compounds/inhibitors are chosen at the SAH binding pocket of DNMT1 and DNMT3A/a of human and mouse. Molecular dynamics (MD) simulations analysis of the potential complexes from docking gives the final clue about the stability of the DNMT–inhibitor complexes. The dynamic picture of the complexes is determined using the time-dependent evolution of the system during the simulation. Changes in free energies for binding (DG) were determined and total energy was decomposed on the basis of per residue contribution. The non-covalent interactions constituting hydrogen bonding, van-der Waals and electrostatic occupancies were monitored throughout the docking and simulations. Moreover, the efficacy of the best found inhibitor is tested by in vitro studies on invasive breast cancer cell line, MDA-MB-231.

2. Materials and methods

146

2.1. In silico investigations

147

2.1.1. Data sets The X-ray crystallographic structures of human DNMT1 (PDB id: 3PTA) [33] and DNMT3a (PDB id: 2QRV) [34] co-crystallized with SAH at resolutions of 3.6 Å and 2.89 Å, respectively, were retrieved from RCSB Protein Data Bank [35]. Subsequently the structure of mouse DNMT1 of 3.25 Å (PDB: 3AV5) [36] cocrystallized with DNA and SAH was obtained from protein data bank. Mouse DNMT1 was used for in silico study because it is used as experimental model in in vivo study. The hetero-atoms including SAH and zinc ion other than those present in active site were edited using chimera software. The double stranded DNA attached to human and mouse DNMT1 were also clipped off. The energy minimization of protein structures was done using steepest descent and conjugate gradient algorithm of 100 steps and step size of 0.02 Å. It followed the addition of polar hydrogen and Gasteiger charges. The set of non-nucleoside inhibitors as described in Fig. 1 was retrieved from PubChem (http://pubchem.ncbi.nlm.nih.gov/) and ChEBI database [37] (www.ebi.ac.uk/chebi). The ligand preparation was done using ‘‘Prepare ligands’’ protocol at Discovery Studio 2.5. The preparation of ligand involved removal of duplicate structure, generation of tautomer, isomers, Lipinski filter, change of ionization state and generating 3D structure.

148

2.1.2. Multiple sequence alignment of DNMTs nucleotide sequence Multiple sequence alignments were carried out for DNMT3A/a, DNMT3B/b amino acid sequences of human and mouse to determine the conserved amino sequence along the active site residues. Amino acid sequences of Uniprot accession number Q9Y6K1, Q9UBC3, O88508 and O88509 were retrieved from Uniprot database [38] located at (http://www.uniprot.org/). The sequence were aligned with an window interface of CLUSTALW [39]. The BLOSUM62 substitution matrix [40] was used, with a gap start penalty of 10 and a gap extend penalty of 0.2.

171

2.2. Docking protocols

181

In order to generalize the ligand and protein conformation on binding accuracy, variable docking algorithms were employed. Each algorithm constitutes alternative way of scoring and treating ligand flexibility keeping protein structure in rigid state in order to reduce conformation search space. The various algorithms used for handling ligand flexibility constituted ‘‘Lamarckian Genetic Algorithm (LGA) (Autodock) [41], monte-carlo conformational search (LigandFit) [42], descriptor matching (Glide_XP) [43] and molecular dynamics simulated annealing (CDOCKER) [44].

182

2.2.1. Autodock AutoDock 3.05 is freely available software availed from Scripps Research Institute. Here, the inhibitors are treated as flexible ligands by modifying their rotatable torsions while the protein template is considered to be a rigid receptor. The minimised protein structures 3PTA, 2QRV and 3AV5 were used as target structures. Grid maps were prepared using auto grid to fix the active site of protein having specific co-ordinates and dimension of 40  40  40 and a resolution of 0.375 Å. Docking parameters were set as: number of individuals in the population (set to 150), maximum number of energy evaluations (set to 2,500,000), maximum number of generations (set to 27,000), and number of hybrid GALS runs (set to 100).

191

Please cite this article in press as: A. Shilpi et al., Mechanisms of DNA methyltransferase–inhibitor interactions: Procyanidin B2 shows new promise for therapeutic intervention of cancer, Chemico-Biological Interactions (2015), http://dx.doi.org/10.1016/j.cbi.2015.03.022

149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170

172 173 174 175 176 177 178 179 180

183 184 185 186 187 188 189 190

192 193 194 195 196 197 198 199 200 201 202 203

CBI 7321

No. of Pages 17, Model 5G

31 March 2015 A. Shilpi et al. / Chemico-Biological Interactions xxx (2015) xxx–xxx

3

Fig. 1. Chemical structure of non-nucleoside inhibitors of DNMTs known till date.

204 205 206 207 208 209 210 211 212 213 214 215 216 217

2.2.2. Glide_XP The protein structure was prepared using the protein preparation module of Schrödinger software. The co-crystallized water molecules were removed. All the selected ligands were assigned an appropriate bond order using the LigPrep 2.4.107 script and converted to .mae format (Maestro, Schrödinger, Inc.) and optimization was carried out by means of the OPLS_2005 force field. The Protein ligand docking studies were performed using Maestro 9.1.107. Parameters having default values were selected and docking was carried out using Glide Extra Precision (XP Glide), version 4.5.19. After the complete preparation of protein and ligand for docking, receptor-grid files were generated. Here van-der Waal radii were scaled of receptor atoms by 1 Å with partial atomic charge of 0.25 for running the grid generation module.

236

2.2.3. CDOCKER CDOCKER is an in-house docking protocol of ‘‘Accelyrs Discovery studio’’. For initial stage MD a softcore potential is used. Each of the structures from the MD run are then located and fully minimized. The solutions are then clustered according to position and conformation and ranked by energy. CHARMm charges are used for the protein structure, i.e., the param19/toph19 parameter set38 using only polar hydrogens. CDOCKER only allows for flexible ligand treatment. Here the docking model constitutes receptor in its rigid state and static protein conformation of biding site is described using 1.0 Å grid and for every point grid, interaction energies of 20 types of probe atoms are calculated. The three dimensional grid is calculated such that radius of 8 Å extend in all directions from any atom in the ligand. Subsequent to simulated annealing conformational search of the flexible ligand, the grid is removed minimization of all atoms of protein–ligand is performed by fixing the coordinates of the protein using the standard all atom potential function with a distance dependent dielectric (RDIE). This interaction energy is taken as the score for the final ligand pose.

237

2.3. LigandFit

238

LigandFit is another docking programme of Accelyrs Discovery Studio. It is based on protein minimization using steepest descent

218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235

239

(gradient D1190:OD2 C1191:H ? EGCG:O7 Q1227:HE22 > EGCG:O3 EGCG:H6 ? M696:SD EGCG:H14 > D1190:OD2 EGCG: H16 ? F1145:O E1168:HE2 ? Prc:O5 C1191:H ? Prc:O11 R1310:HH22 ? Prc:O11 T1528:HG1 ? Prc:O14 Prc:H16 ? E698:OE1 Prc:H23 ? D1190:OD2 Prc:H31 ? G1577:O Prc:H32 ? D701:OD1 Prc:H33 ? D701:OD1

1.8 2.1 2.1 2.3 1.9 2.1 1.8 1.9 2 2.4 2.3 2.1 2.9 1.9 2.2 2.2 2.1 2.2 1.7 1.8 1.6

SAH

9.69

E660,F636,D637,W889,T641,S888,S659,R887,P705, V683,D682,V661,L726

EGCG

10.13

E660,S665,G638,R887,L884,G703,E752,F636,P705, V683,L726,S704,V661

Procyanidin B2 (Prc)

11.53

N707,P705,V683,V661,S659,F636,G681,R887,R684,D682, R883,G722,L726,T723

T641:HG1 ? SAH:OXT V863:H ? SAH:N1 W889:H ? SAH:OXT SAH:HO ? E660:OE1 SAH:H1 ? E660:OE2 SAH:H3 ? D682:OD1 S665:HG ? EGCG:O8 R887:HH22 > EGCG:O8 EGCG:H5 ? E752:OE1 EGCG:H5 ? E660:OE2 EGCG: H13 ? L884:O EGCG:H15 ? E660:OE1 V683:H > Prcd:O20 N707:HD22 ? Prc:O12 N707:HD22 > Prc:O14 R887:HH21 ? Prc:O7 Prc:H23 ? D682:OD1 Prc:H30 ? G722:O Prc:H31 ? G722:O Prc:H33 ? S659:OG Prc:H34 ? D682:OD2

2A 2.4 1.9 2A 2.1 2.4 2.1 2.1 2 2.2 2 2.2 1.8 2.1 2.1 2.3 1.7 2.3 2.4 2.3 2.2

SAH

10.35

E1171,N1580,G1225,G1150,F1148,S1149,A1581,D1146, V1582,L1154,E1269,G1153,G1152,C1151,V1147,I1170, P1228,C1194,N1195,E1192,D1193,L1250,M1172

2.1 2 2.2 2.1 2.1 2.2 2.1 2.1 3.6 1.8 2.4

EGCG

11.0

C1194,P1228,W1173,F1148,E1192,A621, N1248,D1193,N1195,L1250,M1172

Procyanidin (Prc)

14.9

C1194,W1173,L1250,D1174,N1196,Q123,P1228,N1580,G1579, C1229,G1150,G1226,S1149,E1171,F1148

G1152:H ? SAH:O G1153:H ? SAH:O L1154:H ? SAH:OXT C1194:H ? SAH:N1 N1580:HD21 ? SAH:O3 V1582:H ? SAH:OXT SAH:HN2 ? S1149:O SAH:HO ? N1171:OE1 SAH:HO ? N1171:OE2 SAH:HN1 ? D1143:OD1 SAH:HN1 ? D1193:OD1 SAH:H1 ? F1148:O M1172:H ? EGCG:O6 C1194:H ? EGCG:O8 EGCG:H15 > D1193:OD1 EGCG:H17 ? N1248:O EGCG:H17-N1248:OD1 EGCG:H18 > D1193:OD2 C1194:H ? Prc:O17 Q1230:HE11 ? Prc:O4 Q1230:HE22 ? Prc:O13 Prc:H16 ? Q1230:OE1 Prc:H24 ? E1171:OE1 Prc:H27 ? F1149:O Prc:H33 ? G1579:O

Human DNMT3a(2QRV)

Mouse_DNMT1 (3AV5)

2.1 1.9 2 2.2 2.5 1.6 2.1 2.3 2.5 1.7 2.1 2 1.9

Please cite this article in press as: A. Shilpi et al., Mechanisms of DNA methyltransferase–inhibitor interactions: Procyanidin B2 shows new promise for therapeutic intervention of cancer, Chemico-Biological Interactions (2015), http://dx.doi.org/10.1016/j.cbi.2015.03.022

CBI 7321

No. of Pages 17, Model 5G

31 March 2015 A. Shilpi et al. / Chemico-Biological Interactions xxx (2015) xxx–xxx

9

Fig. 5. Depiction of interaction of Procyanidin B2 with (a) 3PTA) (yellow), (b) 2QRV (cyan), (c) 3AV5 (purple) via hydrogen bonds defining specific length. The hydrogen bonds and their respective bond lengths in Å have been shown. The p-cation interactions have been shown by solid orange line. The figure is produced with Accelrys discovery studio. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.) 522 523 524

Finally, DNMT–inhibitor (SAH, EGCG and procyanidin B2) complexes exhibiting higher binding score were selected and subjected to molecular dynamics simulations in explicit aqueous solution.

3.1.4. Molecular dynamics simulation of DNMT–inhibitor complexes To authenticate the docking results and decipher efficacy to inhibiting DNMTs activity, different molecular dynamics

Please cite this article in press as: A. Shilpi et al., Mechanisms of DNA methyltransferase–inhibitor interactions: Procyanidin B2 shows new promise for therapeutic intervention of cancer, Chemico-Biological Interactions (2015), http://dx.doi.org/10.1016/j.cbi.2015.03.022

525 526 527

CBI 7321

No. of Pages 17, Model 5G

31 March 2015 10 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558

A. Shilpi et al. / Chemico-Biological Interactions xxx (2015) xxx–xxx

simulations were implemented. In order to maintain proper orientation of ligand distance, restraints were applied to inhibitors in the initial few picoseconds (ps) and then whole complexes were allowed to move freely. The docked conformations were analysed by examining their relative total energy scores, protein backbone root mean square deviation (RMSD), total hydrogen bonds, vander Waals interaction, electrostatic interaction and root mean square fluctuation (RMSF) of active site residues. Here, the stability of DNMT-inhibitor complex was determined in terms of total energy (kinetic Ek + potential Ep) at the given temperature of 310 K and pressure of 1 atm. We plotted the fluctuation in total energy as a function of constant time gaps. The average total energy (kJ/mol) for EGCG and procyanidin B2 was very high at SAH binding pocket of 3PTA, 2QRV and 3AV5. In Fig. 6 it is evident that both the inhibitors oscillated with nearly same frequency of 1.718  106, 6.868  105 and 4.071  106 kJ/mol. The RMSD of each protein relative to binding of respective inhibitors (SAH, EGCG and procyanidin B2) was calculated to monitor the stability of each trajectory. Here, we mainly focussed on the active site residues of DNMTs. The stability of protein–inhibitor was analysed by aligning heavy atoms of complex to the crystal structures of proteins using the mass-weighted least square fitting method. The RMSD plot exhibited the structures which are stable during the course of MD simulations. The RMSD plot unveiled an increase in deviation at first 200 ps of the production phase. This is because the equilibration phase was performed with restraints on complex, while in production phase the restraints were released. From Fig. 7 one can easily watch that the inhibitor procyanidin B2 attains equilibrium after 500 ps and on average comparatively lesser RMSD of 3.0 Å, 1.6 Å and 2.4 Å is attained on interaction with

3PTA, 2QRV and 3AV5 respectively. Thus from this observation it its evident that procyanidin B2 forms more stable complex than SAH and EGCG with DNMTs. The stability of an enzyme–ligand complex is characterised by ligand binding mode inside the active site pocket of the enzyme and the binding force includes strong hydrogen bonds, electrostatic and van der Waals interactions. The intermolecular hydrogen bond plots between enzyme and inhibitor is shown in Fig. 8. On an average procyanidin B2 forms higher number of hydrogen bonds than SAH and EGCG with the cognate donor/receptor in the vicinity of the binding pocket of DNMTs. It forms an average of 6.5 on interaction with 3PTA while 5.2 hydrogen bonds are formed when ligated to 2QRV and 3AV5. Further, non-bonding interaction constituting Columbic function and Lennard–Jones potential function were employed to calculate electrostatic and van-der Waals interactions, respectively with cut off distance of 9 Å. On evaluation of both the parameters it has been identified that van-der Waals interaction take over the electrostatic interaction, thus favouring the interaction of inhibitors mainly to procyanidin B2 at the active site pocket. The average van-der Waals energy (DEvdw) is identified to 367.743, 296.80 and 330.101 kJ/mol on interaction with 3PTA, 2QRV and 3AV5. Thus from the above findings it can be inferred that hydrogen bonds and van der Waals energy dominates in the total binding energy profile for stabilising the protein–inhibitor complex. The residue flexibility of bound DNMT-inhibitor complexes was examined by analysing the RMSF of the Ca atoms of each residue. The RMSF plot for different protein complexes has shown the flexible regions of the systems; however plot clearly displayed minimum fluctuation around the active site residues (Fig. 9a–c). The amino acid residues exhibit an average fluctuation of 0.98, 0.69,

Fig. 6. Total energy at each ps on interaction of SAH (red), EGCG (blue) and Procyanidin B2 (green) with (a) 3PTA, (b) 2QRV, (c) 3AV5 in kJ/mol. Total energy has been identified to be elevated for EGCG and procyanidin B2 on interaction with DNMT1 in human and mouse and DNMT3A human. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Please cite this article in press as: A. Shilpi et al., Mechanisms of DNA methyltransferase–inhibitor interactions: Procyanidin B2 shows new promise for therapeutic intervention of cancer, Chemico-Biological Interactions (2015), http://dx.doi.org/10.1016/j.cbi.2015.03.022

559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589

CBI 7321

No. of Pages 17, Model 5G

31 March 2015 A. Shilpi et al. / Chemico-Biological Interactions xxx (2015) xxx–xxx

11

Fig. 7. RMSD plot with respect to time in ps on binding of SAH (red), EGCG (blue) and Procyanidin B2 (green) with (a) 3PTA, (b) 2QRV and (c) 3AV5 in Å. RMSD is calculated for heavy atoms with reference to their respective orientation in the crystal structures. Procyanidin B2 exhibit least deviation on interaction with DNMT and forms most stable complex as compared to SAH and EGCG. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

590 591 592 593 594 595 596 597 598 599 600

601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616

and 1.1 Å in case of procyanidin B2 while 1.2, 0.8 and 1.13 Å in case of EGCG, on interaction with 3PTA, 2QRV and 3AV5, respectively. The average fluctuation of amino acid residues on interaction of SAH with these proteins has been found to be 1.3, 1.0 and 1.5 Å, respectively. From the above findings we can infer that procyanidin B2 forms more stable complex because the residues fluctuation is comparatively lesser than others. Thus, stability of procyanidin B2–DNMT complexes has been established by total energy, RMSD of protein backbone, noncovalent interactions and RMSF of active site residues measurements. 3.1.5. Thermodynamic evaluation of DNMT–inhibitor complexes Absolute free energies of binding were evaluated using MM-PBSA method in order to gain insight into continuous spectrum of binding energy of hDNMT1, DNMT3A and mDNMT1 with respect to SAH, EGCG and procyanidin B2. In this method the interaction and solvation energy is computed for complex, receptor and ligand in order to investigate average binding free energy. The detailed binding energy of protein–inhibitor complexes have been depicted in Fig. 10a–c. The free energy of binding has been evaluated with respect to SAH at active site pocket of DNMTs. It is evident from Fig. 10 that procyanidin B2 has highest binding efficiency for DNMTs (3PTA, 2QRV and 3AV5). The binding energies of the 3PTA, 2QRV and 3AV5 with respect to procyanidin B2 are – 16.64, –15.06 and –17.29 kcal/mol respectively. The binding efficiency of EGCG for the respective proteins has also been identified to be greater than SAH. The discrepancy in binding energy of

protein–inhibitor complexes implicates that there are various mode of interactions From Fig. 10, it is evident that the highest binding energy for procyanidin B2-DNMT complexes are consequence of gaseous phase electrostatic and van der Waals interactions. The nonpolar solvation energy (DGnp) also favours binding affinity in active site pockets of enzyme. The relatively lower value of non-polar solvation energies indicates the closeness and integrity of packing of cavity regions. Moreover, the parameters like entropy (TDS) and polar solvation energy which is unfavourable for binding of inhibitors is identified to comparatively less for procyanidin B2 as compared to SAH and EGCG. Thus the calculations of free energy of binding of DNMT–inhibitor complexes elucidate that procyanidin B2 may be a novel inhibitor against DNMTs. Thereafter, the detailed mechanism of interaction was decomposed into inhibitor–residue pairs in order create spectrum of inhibitor–residue interaction.

617

3.1.6. Binding spectrum of residues at active site pocket of DNMTs In order to obtain the detailed thermodynamic description of contribution from amino acid residue to the free energy of binding, the interaction energies were further decomposed to contributions of individual residues through MM-GBSA script in the AMBER 12 suite. The method of residue decomposition aids in understanding the atomistic detail of mechanism of residue–inhibitor interactions. The detailed study of contribution of residues to binding energy for SAH, EGCG and procyanidin B2 with respect to 3PTA, 2QRV and 3AV5 has been depicted in Fig. 11(a–c). Overall the major

633

Please cite this article in press as: A. Shilpi et al., Mechanisms of DNA methyltransferase–inhibitor interactions: Procyanidin B2 shows new promise for therapeutic intervention of cancer, Chemico-Biological Interactions (2015), http://dx.doi.org/10.1016/j.cbi.2015.03.022

618 619 620 621 622 623 624 625 626 627 628 629 630 631 632

634 635 636 637 638 639 640 641 642 643

CBI 7321

No. of Pages 17, Model 5G

31 March 2015 12

A. Shilpi et al. / Chemico-Biological Interactions xxx (2015) xxx–xxx

Fig. 8. Observation of intermolecular hydrogen bond in Å between SAH (red), EGCG (blue), Procyanidin B2 (green) with active site residues of (a) 3PTA, (b) 2QRV and (c) 3AV5. On an average Procyanidin B2 forms higher number of hydrogen bonds with the respective enzymes. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 9. Root-mean-squared fluctuations (RMSF) of backbone atoms (Å) of (a) 3PTA, (b) 2QRV and (c) 3AV5 on binding to SAH (red), EGCG (blue) and Procyanidin B2. Active site residues for all protein–ligand complexes exhibit least fluctuation, exhibiting the stability of the complexes. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Please cite this article in press as: A. Shilpi et al., Mechanisms of DNA methyltransferase–inhibitor interactions: Procyanidin B2 shows new promise for therapeutic intervention of cancer, Chemico-Biological Interactions (2015), http://dx.doi.org/10.1016/j.cbi.2015.03.022

CBI 7321

No. of Pages 17, Model 5G

31 March 2015 A. Shilpi et al. / Chemico-Biological Interactions xxx (2015) xxx–xxx

13

Fig. 10. Energy components (kcal, mol) for the binding of SAH (blue), EGCG (brown) and procyanidin B2 (green) at binding pocket of (a) 3PTA (b) 2QRV and 3AV5. DEele, electrostatic energy in the gas phase; DEvdw, van der Waals energy; DGnp, nonpolar solvation energy; DGpb, polar solvation energy, TDS, total entropy contribution; DGtotal = DEele + DEvdw + DEint + DGpb; DG = DGtotal  TDS. Error bars in green solid line indicates the difference. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

665

interaction at active site of pocket of 3PTA is contribution of F1145, E1168, M1169, E1189 and C1191. The average binding energy of these residues is greater than 1 kcal/mol. Similarly, in case of mouse DNMT1 (3AV5), the residues F1148, E1171, M1172, C1194 and C1248 contributes to the elevation in binding energy. From the Fig. 11a and c it is evident that procyanidin B2 has higher interaction with these residues as compared to SAH and EGCG. Moreover, among all, the cysteine residue offers highest binding energy on interaction with procyanidin B2. C1191 and C1194 contribute to the binding energy of 5.10 and 4.21 kcal/mol to the interaction of procyanidin B2 with 3PTA and 3AV5 respectively. In case of DNMT3A, the residues F636, D637, E660, D684 and W889 imparts to the elevation in binding of inhibitors in the catalytic pocket of the enzyme. The D684 residue has higher contribution to the interaction with procyanidin B2 and the average binding energy is identified to be 4.01 kcal/mol. This decomposition of free binding energy (DG) per residue is consequence of van der Waals (DEvdw) energy, the sum of electrostatic and the polar solvation energy and the non-polar solvation (DGnp) energy. The change of entropy parameter is not included. Further ahead, the in vitro studies were carried out to affirm the role of procyanidin B2 as a potent DNMT inhibitor.

666

3.2. In vitro analyses

3.2.2. Upregulation of DNMT target and DNMTs genes by EGCG and procyanidin B2 To further validate the DNMT inhibitory activity, we examined the effect of procyanidinB2 and EGCG on the expression of DNMT target genes (E-cadherin, Maspin and BRCA1) in MDA-MB-231 cells. Our result indicates that the procyanidin B2 treatment more effectively upregulates the expression of E-cadherin, Maspin and BRCA1 as compared to EGCG. This apparently reveals that procyanidin B2 inhibition of DNMTs causes upregulation of these genes (Fig. 13a). Moreover, the expression of the DNMT genes; DNMT1, DNMT3A and DNMT3B were also enhanced on treatment with these polyphenols (Fig. 13a). The mRNA levels of these genes were identified to be significant (n = 3, mean ± S.D.) between EGCG and procyanidin B2 treated groups. The p-value was significant at 0.05.

667

3.2.1. Effect of EGCG and procyanidin B2 on DNMTs activity The inhibitors, EGCG and procyanidin B2 were used to construct dose–response curve in terms of IC50 value at varying concentration of drugs. The nuclear extracts from MDA-MB-231 cells were incubated with increasing concentrations of EGCG and procyanidin B2 (1, 2.5, 5, 10 and 15 lM). We observed dose dependent growth inhibition of DNMTs on incubation with various concentration EGCG and procyanidin B2. The IC50 was determined under identical

3.2.3. EGCG and procyanidin B2 are non-toxic for normal cells The cytotoxicity analysis examined the toxic nature of EGCG and ProcyanidineB2 towards normal keratinocytes (HaCat) and triple negative breast cancer cells (MDA-MB231) in terms of percentage of cell viability. From the Fig. 14, it is evident that a significant decrease in cell viability was seen with an increasing concentration of EGCG and procyanidin B2 in MDA-MB231 cells. However, these inhibitors did not elicit any lethal effect on normal cells. The sub-

644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664

668 669 670 671 672 673 674

assay conditions. The IC50 of EGCG is identified to be 9.36 ± 1.02 lM, while that of procyanidin B2 is 6.88 ± 0.64 lM (Fig. 12). Thus, from above analysis it can be concluded that procyanidin B2 is effective successfully inhibiting DNMTs at lower dose of drug concentration as compared to EGCG. The difference in IC50 between EGCG and procyanidin B2 was identified to be significant (n = 3, mean ± S.D.). The p-value was found to be significant at 0.05 (p = 0.023).

675

683

698

Please cite this article in press as: A. Shilpi et al., Mechanisms of DNA methyltransferase–inhibitor interactions: Procyanidin B2 shows new promise for therapeutic intervention of cancer, Chemico-Biological Interactions (2015), http://dx.doi.org/10.1016/j.cbi.2015.03.022

676 677 678 679 680 681 682

684 685 686 687 688 689 690 691 692 693 694 695 696 697

699 700 701 702 703 704 705

CBI 7321

No. of Pages 17, Model 5G

31 March 2015 14

A. Shilpi et al. / Chemico-Biological Interactions xxx (2015) xxx–xxx

Fig. 11. Decomposition of DG on a per-residue basis for the SAH (blue), EGCG (brown) and procyanidin B2 (green) at the binding pocket of (a) 3PTA (b) 2QRV and (c) 3AV5. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

Fig. 12. Depiction of log of dose–response plot in terms of percentage decrease in DNMT activity against increasing log of concentration of procyanidin B2 and EGCG. The IC50 of procyanidin B2 and EGCG are found to be 6.88 ± 0.647 lM and 9.36 ± 1.02, respectively. This clearly demonstrates that procyanidin B2 is more active in inhibiting DNMTs. Data are expressed as mean ± S.D., n = 3, p < 0.05.

706 707 708 709 710 711 712

lethal concentration (LC50) of EGCG and procyanidin B2 was determined to be 200 and 150 lM, respectively, in MDA-MB-231 cells. In contrast, at LC50 of EGCG and procyanidin B2, the HaCat cells were found to induce no such cytotoxic phenomena. SAH, being an endogenous biochemical product of the methylation reaction has minimal cytotoxic effect on MDA-MB 231 cells. The difference in LC50 between EGCG and procyanidin B2 was identified to be

significant (n = 3, mean ± S.D.). The p-value was found to be significant at 0.05 (p = 0.01).

713

4. Discussion and conclusions

715

Several studies have documented the apoptosis inducing effect of polyphenols like catechins and procyanidins indicating their anti-cancer potential and chemo therapeutic application [52,53]. However, the mystery behind the molecular targets on cell killing effect of these polyphenolic groups or their effect on epigenetic molecular marks, like DNA methylation and manipulators, like DNMTs was not resolved prior to this work. We are acquainted with the fact that DNMTs are responsible for hypermethylation of various genes, imposing a genotoxic effect including tumour suppressor gene during tumour development and cancer progression [54]. Thus, we sought to explore the factors dictating the selectivity of inhibitors binding to DNMT1 and DNMT3A/a. Docking, molecular dynamics simulation and free energy analyses were carried out in order to unravel the potential of non-nucleoside inhibitors known till date. We found that binding affinity of EGCG was highest among all, successfully inhibiting DNMTs in both human and mouse. The docking and simulation analysis illustrate that EGCG–DNMT complex is energetically favoured and the finding is apparently consistent with an in vitro analysis. Further ahead, EGCG analogues, mainly polyphenolic groups were screened in quest of identification of novel inhibitors. In this second category, procyanidin B2–3, 30 -di-O-gallate moiety presented itself valid and the most promising inhibitor having strong correlation with the active site residues of DNMTs. The selection of procyanidin B2 is of higher negative binding energy and greater selectivity towards DNMTs. Our results demonstrate that binding geometry, driven by hydrogen bond and van-der Waal energy

716

Please cite this article in press as: A. Shilpi et al., Mechanisms of DNA methyltransferase–inhibitor interactions: Procyanidin B2 shows new promise for therapeutic intervention of cancer, Chemico-Biological Interactions (2015), http://dx.doi.org/10.1016/j.cbi.2015.03.022

714

717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742

CBI 7321

No. of Pages 17, Model 5G

31 March 2015 A. Shilpi et al. / Chemico-Biological Interactions xxx (2015) xxx–xxx

15

Fig. 13. Real time RT-PCR analysis of (a) E-cadherin, Maspin, BRCA1, and (b) DNMT1, DNMT3A, DNMT3B gene expression after treatment with EGCG and procyanidin B2. The mRNA level of both DNMT target and DNMT genes are upregulated more in case of procyanidin B2 than EGCG. Data are expressed as mean ± S.D., n = 3, p < 0.05.

Fig. 14. Effect of (a) SAH, (b) EGCG and (c) Procyanidin B2 on cell viability and growth. MDA-MB231 and HaCat cells were treated with indicated concentrations of (a) SAH, (b) EGCG and (c) Procyanidin B2 in lM for 24 h. Cell viability was determined by MTT assay. Data are represented as the mean ± SD of three different observations. EGCG and procyanidin B2 exhibits LC50 of 200 lM and 150 lM, respectively, while SAH has been identified having almost negligible cell growth inhibitory effect for MDA MB 231. However, ECGC and Procyanidin B2 show no cytotoxic effect on HaCat cells. Data are expressed as mean ± S.D., n = 3, p < 0.05. 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759

dominates in enhancement of total binding energy. Electrostatic interaction occupancy is comparatively feeble along the interface of protein–inhibitor complex. Molecular dynamics simulation of protein–ligand complexes fortified the docking analysis. The stability of the inhibitor inside the binding pocket has been substantiated by total energy, RMSD and RMSF data. Furthermore, the analysis of free energy of binding by MM-PBSA for SAH, EGCG and procyanidin B2 with DNMTs established that procyanidin B2 has highest efficacy for the catalytic pocket. Moreover, the detailed thermodynamic description of residue contribution to the free energy of binding affirms the intimate interaction with active site residues. In addition to the results rationalising in silico observed selectivity, in vitro experimental analyses also revealed the potential of procyanidin B2 as an effective inhibitor for diminishing DNMTs activity. The effect of procyanidin B2 in inhibiting DNMTs was evaluated following direct as well as indirect approach. In direct

approach, procyanidin B2 was directly used as an inhibitor against DNMTs in nuclear extract of the cells and DNMT activity was noted to be declined with respect to control. Moreover, procyanidin B2 elicits a greater reduction of DNMT activity at concentration of 6.88 ± 0.647 lM as compared to EGCG. The indirect approach deals with reactivation of DNMT target genes on application of procyanidin B2. Previously, it has been documented that E-cadherin [55], Maspin [56] and BRCA1 [57] are epigenetically inactivated in breast cancer due to aberrant cytosine methylation in their promoter regions. Evidences also report that inhibition of DNA methylation with 5-Aza-20 -deoxycytidine (AZA) could restore the Ecadherin and Maspin gene expression in this cell line [56,58]. Moreover, in prostrate cancer cells it has been reported that BRCA1 can be reactivated on treatment with AZA [23]. Based on this fact, we sought to examine whether EGCG and procyanidin B2 could be able to restore the expression of these DNMT target genes as that of AZA. Corroborating our in silico analysis and

Please cite this article in press as: A. Shilpi et al., Mechanisms of DNA methyltransferase–inhibitor interactions: Procyanidin B2 shows new promise for therapeutic intervention of cancer, Chemico-Biological Interactions (2015), http://dx.doi.org/10.1016/j.cbi.2015.03.022

760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776

CBI 7321

No. of Pages 17, Model 5G

31 March 2015 16

A. Shilpi et al. / Chemico-Biological Interactions xxx (2015) xxx–xxx

809

in vitro DNMT activity inhibition measurement, we found that both EGCG and procyanidin B2 lead to reactivation of E-cadherin, Maspin and BRCA1 gene at transcription level. Of note, procyanidin B2 could more efficiently upregulate the DNMT target gene expression relative to EGCG. This might be due to more affinity of procyanidin B2 for DNMTs, thus inhibiting DNMTs to a greater extent than EGCG. However, this enzymatic inhibition impels the elevated expression of DNMTs. Consequently, among the non-nucleoside inhibitors procyanidin B2 can be considered to be more effective in reducing the DNMT activity and hence can be used to decrease the methylation level. One of the foremost concerns for clinical application of any anticancer drug relies on its clinical toxicity. So, after searching a potential DNMT inhibitor as procyanidin B2, our next attempt was to examine its cytotoxic nature towards normal cells in contrast to breast cancer cells. Both EGCG and procyanidinB2 were found to elicit extensive cytotoxic effect in highly invasive triple negative MDA-MB-231 breast cancer cells. While EGCG reduced 50% cell viability at 200 lM concentration, procyanidin B2 was effective at comparatively lower concentration of 150 lM in 24 h. Conversely, the same LC50 of EGCG and procyanidinB2 treatment exhibited no cytotoxic activity to normal keratinocytes (HaCat). In contrast to conventional chemotherapeutic drugs, procyanidin B2 is non-toxic in nature. Additionally, it is natural and a dietary component with substantial anticancer effects on breast cancer cells. In conclusion, we have unravelled the role of procyanidin B2 as an epigenetic modulator which precisely targets DNMTs and reverses the silencing of tumor suppressor genes. The outcome of this investigation holds procyanidin B2 as a promising inhibitor against cancer targeting the enzyme DNMTs. Further, by executing experiments in animal models and clinical settings, it may be recommended for incorporation in the list of compounds in chemoprevention of breast cancer.

810

Author contributions

811

814

The project was conceived by S.K.P. and A.S. The experiments were performed by A.S., S.P. and S.K. The manuscript was written by A.S. and S.K.P. through contributions of all authors. All authors have given consent to the final version of the manuscript.

815

Notes

777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808

812 813

816

The authors declare no competing financial interest.

817

Acknowledgements

818

827

A. Shilpi, D. Sengupta, S. Kar, M. Deb and S. K. Rath are thankful to NIT-Rourkela for granting those fellowships under the Institute Research Scheme. S. Parbin and N. Pradhan is thankful to DST, Govt. of India for INSPIRE fellowship. M Rakshit is thankful to NITRourkela for granting her postdoctoral research associate position and fellowship. We are thankful to Professor Tapash C. Ghosh, Bose Institute, Kolkata for helping us for using and analysis with discovery Studio. We would also like to acknowledge Dr. Vinod Devraji who helped with application manuals of Schrodinger software.

828

References

819 820 821 822 823 824 825 826

829 830 831 832 833 834 835

[1] C. Bock, Analysing and interpreting DNA methylation data, Nat. Rev. Genet. 13 (2012) 705–719. [2] S.K. Patra, S. Bettuzzi, Epigenetic DNA-(cytosine-5-carbon) modifications: 5aza-20 -deoxycytidine and DNA-demethylation, Biochemistry (Mosc.) 74 (2009) 613–619. [3] W. Reik, Stability and flexibility of epigenetic gene regulation in mammalian development, Nature 447 (2007) 425–432.

[4] S.K. Patra, M. Deb, A. Patra, Molecular marks for epigenetic identification of developmental and cancer stem cells, Clin. Epigenetics 2 (2011) 27–53. [5] A. Meissner, Epigenetic modifications in pluripotent and differentiated cells, Nat. Biotechnol. 28 (2010) 1079–1088. [6] D.P. Barlow, Genomic imprinting: a mammalian epigenetic discovery model, Annu. Rev. Genet. 45 (2011) 379–403. [7] T.H. Bestor, The host defence function of genomic methylation patterns, Novartis Found. Symp. 214 (1998) 187–195 (discussion 195–189, 228–132). [8] S.K. Patra, M. Szyf, DNA methylation-mediated nucleosome dynamics and oncogenic Ras signaling: insights from FAS, FAS ligand and RASSF1A, FEBS J. 275 (2008) 5217–5235. [9] S.K. Patra, A. Patra, F. Rizzi, T.C. Ghosh, S. Bettuzzi, Demethylation of (Cytosine5-C-methyl) DNA and regulation of transcription in the epigenetic pathways of cancer development, Cancer Metastasis Rev. 27 (2008) 315–334. [10] P.A. Jones, S.B. Baylin, The epigenomics of cancer, Cell 128 (2007) 683–692. [11] S.K. Patra, A. Patra, H. Zhao, R. Dahiya, DNA methyltransferase and demethylase in human prostate cancer, Mol. Carcinog. 33 (2002) 163–171. [12] F. Lyko, R. Brown, DNA methyltransferase inhibitors and the development of epigenetic cancer therapies, J. Natl Cancer Inst. 97 (2005) 1498–1506. [13] X. Cheng, R.M. Blumenthal, Mammalian DNA methyltransferases: a structural perspective, Structure (London, England: 1993) 16 (2008) 341–350. [14] K.D. Robertson, A.P. Wolffe, DNA methylation in health and disease, Nat. Rev. Genet. 1 (2000) 11–19. [15] S. Kar, M. Deb, D. Sengupta, A. Shilpi, S. Parbin, J. Torrisani, S. Pradhan, S. Patra, An insight into the various regulatory mechanisms modulating human DNA methyltransferase 1 stability and function, Epigenetics 7 (2012) 994–1007. [16] A. Hermann, H. Gowher, A. Jeltsch, Biochemistry and biology of mammalian DNA methyltransferases, Cell. Mol. Life Sci. 61 (2004) 2571–2587. [17] J. Yoo, J.L. Medina-Franco, Homology modeling, docking and structure-based pharmacophore of inhibitors of DNA methyltransferase, J. Comput. Aided Mol. Des. 25 (2011) 555–567. [18] T.H. Bestor, The DNA methyltransferases of mammals, Hum. Mol. Genet. 9 (2000) 2395–2402. [19] J.P. Issa, P.M. Vertino, J. Wu, S. Sazawal, P. Celano, B.D. Nelkin, S.R. Hamilton, S.B. Baylin, Increased cytosine DNA-methyltransferase activity during colon cancer progression, J. Natl Cancer Inst. 85 (1993) 1235–1240. [20] T. Suzuki, R. Tanaka, S. Hamada, H. Nakagawa, N. Miyata, Design, synthesis, inhibitory activity, and binding mode study of novel DNA methyltransferase 1 inhibitors, Bioorg. Med. Chem. Lett. 20 (2010) 1124–1127. [21] D.V. Santi, A. Norment, C.E. Garrett, Covalent bond formation between a DNAcytosine methyltransferase and DNA containing 5-azacytosine, Proc. Natl. Acad. Sci. U.S.A. 81 (1984) 6993–6997. [22] F. Sorm, A. Piskala, A. Cihak, J. Vesely, 5-Azacytidine, a new, highly effective cancerostatic, Experientia 20 (1964) 202–203. [23] A. Patra, M. Deb, R. Dahiya, S.K. Patra, 5-Aza-20 -deoxycytidine stress response and apoptosis in prostate cancer, Clin. Epigenetics 2 (2011) 339–348. [24] J.H. Beumer, R.A. Parise, E.M. Newman, J.H. Doroshow, T.W. Synold, H.J. Lenz, M.J. Egorin, Concentrations of the DNA methyltransferase inhibitor 5-fluoro20 -deoxycytidine (FdCyd) and its cytotoxic metabolites in plasma of patients treated with FdCyd and tetrahydrouridine (THU), Cancer Chemother. Pharmacol. 62 (2008) 363–368. [25] M.Z. Fang, Y. Wang, N. Ai, Z. Hou, Y. Sun, H. Lu, W. Welsh, C.S. Yang, Tea polyphenol ()-epigallocatechin-3-gallate inhibits DNA methyltransferase and reactivates methylation-silenced genes in cancer cell lines, Cancer Res. 63 (2003) 7563–7570. [26] S. Jagadeesh, S. Sinha, B.C. Pal, S. Bhattacharya, P.P. Banerjee, Mahanine reverses an epigenetically silenced tumor suppressor gene RASSF1A in human prostate cancer cells, Biochem. Biophys. Res. Commun. 362 (2007) 212–217. [27] Z. Liu, Z. Xie, W. Jones, R.E. Pavlovicz, S. Liu, J. Yu, P.K. Li, J. Lin, J.R. Fuchs, G. Marcucci, C. Li, K.K. Chan, Curcumin is a potent DNA hypomethylation agent, Bioorg. Med. Chem. Lett. 19 (2009) 706–709. [28] C. Arce, B. Segura-Pacheco, E. Perez-Cardenas, L. Taja-Chayeb, M. Candelaria, A. Duennas-Gonzalez, Hydralazine target: from blood vessels to the epigenome, J. Transl. Med. 4 (2006) 10. [29] A. Villar-Garea, M.F. Fraga, J. Espada, M. Esteller, Procaine is a DNAdemethylating agent with growth-inhibitory effects in human cancer cells, Cancer Res. 63 (2003) 4984–4989. [30] B.H. Lee, S. Yegnasubramanian, X. Lin, W.G. Nelson, Procainamide is a specific inhibitor of DNA methyltransferase 1, J. Biol. Chem. 280 (2005) 40749– 40756. [31] L. Isakovic, O.M. Saavedra, D.B. Llewellyn, S. Claridge, L. Zhan, N. Bernstein, A. Vaisburg, N. Elowe, A.J. Petschner, J. Rahil, N. Beaulieu, F. Gauthier, A.R. MacLeod, D. Delorme, J.M. Besterman, A. Wahhab, Constrained (l-)-Sadenosyl-L-homocysteine (SAH) analogues as DNA methyltransferase inhibitors, Bioorg. Med. Chem. Lett. 19 (2009) 2742–2746. [32] O.M. Saavedra, L. Isakovic, D.B. Llewellyn, L. Zhan, N. Bernstein, S. Claridge, F. Raeppel, A. Vaisburg, N. Elowe, A.J. Petschner, J. Rahil, N. Beaulieu, A.R. MacLeod, D. Delorme, J.M. Besterman, A. Wahhab, SAR around (l)-S-adenosylL-homocysteine, an inhibitor of human DNA methyltransferase (DNMT) enzymes, Bioorg. Med. Chem. Lett. 19 (2009) 2747–2751. [33] J. Song, O. Rechkoblit, T.H. Bestor, D.J. Patel, Structure of DNMT1-DNA complex reveals a role for autoinhibition in maintenance DNA methylation, Science (New York, N.Y.) 331 (2011) 1036–1040. [34] D. Jia, R.Z. Jurkowska, X. Zhang, A. Jeltsch, X. Cheng, Structure of Dnmt3a bound to Dnmt3L suggests a model for de novo DNA methylation, Nature 449 (2007) 248–251.

Please cite this article in press as: A. Shilpi et al., Mechanisms of DNA methyltransferase–inhibitor interactions: Procyanidin B2 shows new promise for therapeutic intervention of cancer, Chemico-Biological Interactions (2015), http://dx.doi.org/10.1016/j.cbi.2015.03.022

836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921

CBI 7321

No. of Pages 17, Model 5G

31 March 2015 A. Shilpi et al. / Chemico-Biological Interactions xxx (2015) xxx–xxx 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962

[35] H.M. Berman, J. Westbrook, Z. Feng, G. Gilliland, T.N. Bhat, H. Weissig, I.N. Shindyalov, P.E. Bourne, The protein data bank, Nucleic Acids Res. 28 (2000) 235–242. [36] K. Takeshita, I. Suetake, E. Yamashita, M. Suga, H. Narita, A. Nakagawa, S. Tajima, Structural insight into maintenance methylation by mouse DNA methyltransferase 1 (Dnmt1), Proc. Natl. Acad. Sci. U.S.A. 108 (2011) 9055– 9059. [37] C. Brooksbank, G. Cameron, J. Thornton, The European bioinformatics Institute’s data resources: towards systems biology, Nucleic Acids Res. 33 (2005) D46–D53. [38] R. Apweiler, A. Bairoch, C.H. Wu, W.C. Barker, B. Boeckmann, S. Ferro, E. Gasteiger, H. Huang, R. Lopez, M. Magrane, M.J. Martin, D.A. Natale, C. O’Donovan, N. Redaschi, L.S. Yeh, UniProt: the universal Protein knowledgebase, Nucleic Acids Res. 32 (2004) D115–119. [39] J.D. Thompson, D.G. Higgins, T.J. Gibson, CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice, Nucleic Acids Res. 22 (1994) 4673–4680. [40] S. Henikoff, J.G. Henikoff, Amino acid substitution matrices from protein blocks, Proc. Natl. Acad. Sci. U.S.A. 89 (1992) 10915–10919. [41] G.M. Morris, D.S. Goodsell, R.S. Halliday, R. Huey, W.E. Hart, R.K. Belew, A.J. Olson, Automated docking using a Lamarckian genetic algorithm and an empirical binding free energy function, J. Comput. Chem. 19 (1998) 1639– 1662. [42] C.M. Venkatachalam, X. Jiang, T. Oldfield, M. Waldman, LigandFit: a novel method for the shape-directed rapid docking of ligands to protein active sites, J. Mol. Graph. Model. 21 (2003) 289–307. [43] R.A. Friesner, J.L. Banks, R.B. Murphy, T.A. Halgren, J.J. Klicic, D.T. Mainz, M.P. Repasky, E.H. Knoll, M. Shelley, J.K. Perry, D.E. Shaw, P. Francis, P.S. Shenkin, Glide: a new approach for rapid, accurate docking and scoring. 1. Method and assessment of docking accuracy, J. Med. Chem. 47 (2004) 1739–1749. [44] G. Wu, D.H. Robertson, C.L. Brooks 3rd., M. Vieth, Detailed analysis of gridbased molecular docking: a case study of CDOCKER-A CHARMm-based MD docking algorithm, J. Comput. Chem. 24 (2003) 1549–1562. [45] S. Pronk, S. Pall, R. Schulz, P. Larsson, P. Bjelkmar, R. Apostolov, M.R. Shirts, J.C. Smith, P.M. Kasson, D. van der Spoel, B. Hess, E. Lindahl, GROMACS 45: a highthroughput and highly parallel open source molecular simulation toolkit, Bioinformatics (Oxford, England) 29 (2013) 845–854. [46] A.W. Schuttelkopf, D.M. van Aalten, PRODRG: a tool for high-throughput crystallography of protein–ligand complexes, Acta Crystallogr. Sect. D Biol. Crystallogr. 60 (2004) 1355–1363.

17

[47] T. Hou, J. Wang, Y. Li, W. Wang, Assessing the performance of the MM/PBSA and MM/GBSA methods. 1. The accuracy of binding free energy calculations based on molecular dynamics simulations, J. Chem. Inf. Model. 51 (2011) 69– 82. [48] P.A. Kollman, I. Massova, C. Reyes, B. Kuhn, S. Huo, L. Chong, M. Lee, T. Lee, Y. Duan, W. Wang, O. Donini, P. Cieplak, J. Srinivasan, D.A. Case, T.E. Cheatham 3rd., Calculating structures and free energies of complex molecules: combining molecular mechanics and continuum models, Acc. Chem. Res. 33 (2000) 889–897. [49] V. Tsui, D.A. Case, Theory and applications of the generalized Born solvation model in macromolecular simulations, Biopolymers 56 (2000) 275–291. [50] A.S. Baldwin Jr., The NF-kappa B and I kappa B proteins: new discoveries and insights, Annu. Rev. Immunol. 14 (1996) 649–683. [51] J. Yoo, J.H. Kim, K.D. Robertson, J.L. Medina-Franco, Molecular modeling of inhibitors of human DNA methyltransferase with a crystal structure: discovery of a novel DNMT1 inhibitor, Adv. Protein Chem. Struct. Biol. 87 (2012) 219– 247. [52] S. Dinicola, A. Cucina, A. Pasqualato, S. Proietti, F. D’Anselmi, G. Pasqua, A.R. Santamaria, P. Coluccia, A. Lagana, D. Antonacci, A. Giuliani, M. Bizzarri, Apoptosis-inducing factor and caspase-dependent apoptotic pathways triggered by different grape seed extracts on human colon cancer cell line Caco-2, Br. J. Nutr. 104 (2010) 824–832. [53] M. Kaur, C. Agarwal, R. Agarwal, Anticancer and cancer chemopreventive potential of grape seed extract and other grape-based products, J. Nutr. 139 (2009) 1806S–1812S. [54] S.K. Patra, Ras regulation of DNA-methylation and cancer, Exp. Cell Res. 314 (2008) 1193–1201. [55] S.J. Nass, J.G. Herman, E. Gabrielson, P.W. Iversen, F.F. Parl, N.E. Davidson, J.R. Graff, Aberrant methylation of the estrogen receptor and E-cadherin 50 CpG islands increases with malignant progression in human breast cancer, Cancer Res. 60 (2000) 4346–4348. [56] M.M. Oshiro, G.S. Watts, R.J. Wozniak, D.J. Junk, J.L. Munoz-Rodriguez, F.E. Domann, B.W. Futscher, Mutant p53 and aberrant cytosine methylation cooperate to silence gene expression, Oncogene 22 (2003) 3624–3634. [57] R. Ben Gacem, M. Hachana, S. Ziadi, S. Ben Abdelkarim, S. Hidar, M. Trimeche, Clinicopathologic significance of DNA methyltransferase 1, 3a, and 3b overexpression in Tunisian breast cancers, Hum. Pathol. 43 (2012) 1731–1738. [58] A. Hurtubise, R.L. Momparler, Evaluation of antineoplastic action of 5-aza-20 deoxycytidine (Dacogen) and docetaxel (Taxotere) on human breast, lung and prostate carcinoma cell lines, Anticancer Drugs 15 (2004) 161–167.

Please cite this article in press as: A. Shilpi et al., Mechanisms of DNA methyltransferase–inhibitor interactions: Procyanidin B2 shows new promise for therapeutic intervention of cancer, Chemico-Biological Interactions (2015), http://dx.doi.org/10.1016/j.cbi.2015.03.022

963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003

Mechanisms of DNA methyltransferase-inhibitor interactions: Procyanidin B2 shows new promise for therapeutic intervention of cancer.

DNA methyltransferases (DNMTs) is a key epigenetic enzyme for pharmacological manipulation and is employed in cancer reprogramming. During past few ye...
5MB Sizes 0 Downloads 10 Views