Research Journal of Chemical Sciences ______________________________________________ ISSN 2231-606X Vol. 4(3), 76-80, March (2014) Res. J. Chem. Sci. International Science Congress Association 76 Inverse QSAR approach and Molecular docking studies to design novel methoxy substituted Chalcones and their Computational Anticancer activity evaluationPachori Kanti, Sharma Sonal, Veerkar Prachi3 and Yadav MukeshDept. of Chemistry, Govt. Nirbhay Singh Patel Science College, Indore, MP, INDIA Dept. of Chemistry, Govt. Holkar Science College, Indore, MP, INDIA Dept. of Pharmaceutical Chemistry, Softvision College, Indore, MP, INDIAAvailable online at: www.isca.in, www.isca.me Received 2nd February 2014, revised 19th February 2014, accepted 16th March 2014Abstract Quantitative structure activity-relationship (QSAR) studies have emerged as promising tool to in silico prediction and optimization of potential bioactive compounds. The purpose of QSAR studies is to save time, cost, and animal toxicity and to support green chemistry. Present studies are efforts to design and identify novel chalcones having high potency and selectivity. Present investigations identify structural insights of methoxy substituted chalcones in Linear and non-linear QSAR models. QSAR studies identify the profound non-linear relationship among structures of methoxy substituted chalcones and their biological activity measures (IC50). It concludes that any structural variation in present class of chalcones would bring a non-linear change in IC50. MLR produced efficient QSAR models (R2 = 0.809). We have designed new candidates employing structure-activity relationship obtained from QSAR models. Descriptor based inverse QSAR approach has been applied in computational modeling of new small molecules. Furthermore, they have been compared with synthesized dataset of methoxy substituted chalcones using molecular docking and ADMET studies. In course of molecular docking studies, newly designed molecules yielded promising results with better binding capacity (Docking rerank Score -103.089) than previously synthesized compounds. Computational pharmacokinetic and pharmacodynamic (ADMET) studies revealed their better intestinal absorption, skin permeability and blood brain barrier penetration. The aim of our work is computational designing of novel methoxy substituted chalcones using QSAR and flexible molecular docking based techniques. Keywords: Inverse QSAR approach, flexible molecular docking, novel methoxy substituted chalcones, anticancer activity. Introduction Medicinal chemists today are facing a serious challenge because of the increased cost and enormous amount of time taken to discover a new drug, and also because of rigorous competition amongst different pharmaceutical companies. Thereby, importance of Computer Aided Drug Design (CADD) and molecular modeling is increasing nowadays. Target specific drug discovery is the need of the hour. Techniques evolved in the post genomic era have given us an opportunity to accelerate discovery process by looking at many cellular processes simultaneously. Advances in computational power, algorithms and modern database mining techniques are accelerating the discovery in science even more. A major goal in current drug design is to develop new ligands with high affinity of binding toward a given protein receptor. Pharmacophore, which is the three dimensional arrangement of essential features that enable a molecule to exert a particular biological effect, is a very useful model for achieving this goal. If the three dimensional structure of the receptor is known, pharmacophore is a complementary tool to standard techniques such as docking, molecular modeling and offshoot of theoretical chemistry and an emerging new science, provides many technical advances in reducing the cost of drug discovery. A recent estimate puts about 8 months and $ 66 million savings for each drug due to use of this advance technology. Molecular modeling and pharmaco-informatics, a new emerging field integrates bioinformatics and chemoinformatics applicable to drug discovery. This has been projected to save an additional 4-8 months time and about $70 million per drug. Quantitative Structure Activity Relationships (QSAR) analyse the correlation between the structural features and the biological activity in order to predict the activity level of new compounds. Using statistical correlation methods, it builds models to predict quantities such as binding affinity, toxicity, or pharmacokinetic parameters of existing or hypothetical molecules. 3D-QSAR analyse the three dimensional forces like hydrogen bonds, metal ligand contacts, polarization effect and the interaction between the electric dipoles. Chalcones are the precursors of flavanoid and isoflavanoid family. Chalcones are -unsaturated ketone containing the reactive ketoethylenic group –CO-CH=CH-. Chalcones itself Research Journal of Chemical Sciences ___________________________________________________________ ISSN 2231-606XVol. 4(3), 76-80, March (2014) Res. J. Chem. Sci. International Science Congress Association 77 and its derivatives are posses wide range of therapeutic activities as antioxidant, cytotoxic,6-8 antibacterial, anti malarial, anti inflammatory10,11, anti – HIV, antifungal, tyrosine kinase inhibitory activity12,13. Chalcones exhibit cancer cell proliferation by various mechanisms as inducing apoptosis; inhibiting tubulin polymerization14,15, inhibiting angiogenesis16,17, anti estrogenic activity, uncouple the mitochondrial respiration and collapse the mitochondrial membrane potential18. Methodology Target Structure Selection and Preparation: In structure-based computational techniques, it is crucial that we retrieve and use appropriate 3D structure of drug target. It is even more recommendable to prepare 3D structure of target towards any missing chemical information in terms of hybridization, bond lengths, bond angles, torsions etc19. Present studies included 3D structure Tubulin-Cholchicin protein bound with stathmin-like domain complex (PDB code: 1SA0) available from protein data bank (PDB) at the Research Collaboratory for Structural Bioinformatics (RCSB). We decided to use this specific structure since we use only B-chain of the protein composed of cofactors, ligands, -chain (a.c) length-451, -chain (b.d) length-445 and E-chain of stathmin length-142. Molecules sketched, cleaned in 3D and saved in MDL SDF file format imported in the workspace. For protein preparation while import select the option always in assign all below in Molegro Virtual Docker (MVD). It assign bonds, assign bond orders and hybridization, create explicit hydrogen, assign charges, detect flexible torsions in ligands, assign tripos atom types. Preparation options (If Missing, Always, Never, Remove) applies to each individual molecule (not each individual bond or atom). Docking Parameters: Molegro Virtual Docker (MVD) sets latest and efficient algorithms named PLANTS and PLP scores to calculate molecular fields and receptor-ligand interactions. PLANTS score was developed by Korb et al20 while combined efforts of Yang et al21 and Gehlhaar et al developed PLP score22,23. We have selected score as MolDock score [GRID] with GRID resolution of 20 A°. Ligand evaluation was applied with internal electrostatic ES, internal H-bond, sp2- sp2 torsions. Binding site coordinates:X- 116.763, Y-90.64, and Z-6.248 within a constraint of 10 A°. Though a receptor-ligand interaction is depicted by a lock and key model arbitrary, it is much more complex and case specific specially when there could be many keys (ligands) opening same lock (receptor) to various extents. Selection of accurate site and most favourable pose is bottle neck step in performing molecular docking24. Inhibitor binding site was identified from literature; in addition, all the available cavities were screened for possible binding. Inverse QSAR Design of novel Chalcones: The novel molecules were designed using inverse QSAR approach wherein we used QSAR models obtained. In this approach the significant descriptors responsible for regulation for IC50 (nM) values chalcones were studied and optimized. The MLR results of QSAR studies were employed to calculate or predict the potential (IC50, nM) of new molecules designed computationally. The tetra-variable model used for designing of novel methoxy substitutes Chalcones, Model-4 Tetra-variable: IC50 (nM) = 849.395 + 232.141(Mor21) - 679.357(Mor29e) - 1023.803(Ele) + 91.605 (Depressant 50) N=26, R=0.809, R=0.773 Perusal reveals that increase in the value of coefficients for descriptors with negative coefficient and decrease in the coefficient value for descriptors with positive coefficient value could enhance the activity of the molecules and yield better molecules with best biological activity. 45 new molecules were designed using descriptor based approach. Results and Discussion QSAR Studies: In the present work novel methoxy substituted chalcones are designed. The molecular descriptors are numerical representation to evaluate and establish the structural activity relationship. The structures of chalcones drawn in ChemSketch, afterword they converted into SMILES data format (Simplified molecular line entry specification). The SMILES data format was used to calculate descriptors using E-Dragon (version 5.4). The 2500 descriptors belonging to various classes were calculated and imported into SARCHITECT evaluation version along with structures and their respective biological activity. The MLR model of QSAR was prepared for methoxy substituted chalcones then significant and internally non-correlated sets of descriptors have been chosen with target size four (tetra-variable) limited to thumb rule. MLR results have been discussed using tri-variable and tetra-variable models with appreciable set of statistical parameters as shown in table 1. To produce the novel methoxy substituted chalcone compounds for anticancer activity, we used approximately 2000-2500 molecular descriptors in QSAR, out of that we divided descriptors in to training and test data sets. Tetra-variable MLR model obtained (figure 1) shows linearity between predicted and observed IC50 (nM) activities. Docking Studies: Docking studies have provided the comparative view for binding capacity of dataset molecules (synthesized) with new molecules designed using QSAR. Docking is applied as evaluation criteria for newly designed molecules. Designed molecule TC-14 shows better binding capacity (Rerank Score -103.089) to that of dataset molecules reported in literature. While molecule No. 15 and 34 docked with rerank score -102.854 and -98.128 respectively. Table 2 presents predicted IC50 (nM) of newly designed chalcones and docking scores. Research Journal of Chemical Sciences ___________________________________________________________ ISSN 2231-606XVol. 4(3), 76-80, March (2014) Res. J. Chem. Sci. International Science Congress Association 78 Table-1 Statistics of multivariate models used in MLR Models R 2 R 2 CV R 2 A S.E. F-STAT Uni-variable Ele 0.335 0.2327 0.308 67.884 12.130 Bi-variable Mor29e+Ele 0.562 0.4512 0.524 56.288 14.774 Tri-variable Mor21u+Mor29e+Ele 0.654 0.5241 0.607 51.098 13.922 Tetra-variable Mor21u+Mor29e+ Ele + Depressant-50 0.809 0.7209 0.773 38.87 22.288 Figure-1 Training set correlation of observed and calculated IC50 nM using tetravariable model of MLR Research Journal of Chemical Sciences ___________________________________________________________ ISSN 2231-606XVol. 4(3), 76-80, March (2014) Res. J. Chem. Sci. International Science Congress Association 79 Table-2 Docking results of newly designed molecules Designed Molecule Code MolDock Score(GRID) Score(GRID) MolDock Score Rerank Score Torsions TC-01 -101.33 -100.597 -86.5799 3 TC-02 -90.8473 -88.6096 -81.7398 3 TC-03 -99.3372 -100.126 -88.5429 3 TC-04 -99.3228 -99.1151 -85.4758 3 TC-05 -95.8902 -92.9692 -85.026 3 TC-06 -95.3092 -96.1909 -85.1984 4 TC-07 -101.141 -100.296 -86.7889 3 TC-08 -93.5568 -94.2729 -82.5531 2 TC-09 -93.5605 -94.2828 -82.5274 2 TC- 10 -91.408 -91.9753 -81.1378 2 TC-11 -93.1726 -93.7984 -82.3324 2 TC-12 -91.409 -92.0253 -81.2265 2 TC-13 -92.0234 -92.4188 -81.3354 2 TC-14 * -119.638 -119.321 -103.089 3 TC-15 -119.625 -119.289 -102.854 3 TC-16 -92.5672 -93.3589 -82.1232 2 TC-17 -91.4407 -92.1835 -81.1593 2 TC-18 -92.0636 -92.9503 -81.904 2 TC-19 -92.0636 -92.9464 -81.8328 2 TC-20 -98.4921 -99.8874 -81.7211 3 TC-21 -98.0289 -99.0462 -84.6618 3 TC-22 -98.137 -99.0713 -83.5998 3 TC-23 -98.1239 -99.0365 -84.5198 3 TC-24 -98.1495 -99.0723 -84.9271 3 TC-25 -95.5168 -96.1614 -83.5972 3 TC-26 -97.0112 -97.2025 -85.8632 3 TC-27 -97.6989 -98.5404 -87.5783 3 TC-28 -93.6295 -94.1854 -82.5108 3 TC-29 -95.9762 -96.5836 -84.9201 3 TC-30 -97.7272 -98.6252 -86.6424 3 TC-31 -132.964 -134.718 -97.0981 8 TC-32 -116.643 -116.631 -86.827 5 TC-33 -110.381 -107.472 -84.3514 2 TC-34 -109.997 -111.291 -98.1283 3 TC-35 -115.75 -115.295 -94.7381 3 TC-36 -105.414 -106.517 -92.0256 3 TC-37 -106.533 -107.605 -90.3713 2 TC-38 -106.457 -103.487 -91.1246 3 TC-39 -97.4048 -98.0125 -84.0592 2 TC-40 -95.4614 -96.1517 -83.9843 2 TC-41 -100.449 -101.212 -82.8946 2 TC-42 -101.094 -102.072 -84.3186 2 TC-43 -102.784 -103.706 -84.6786 2 TC-44 -101.973 -101.563 -81.7831 2 TC-45 -95.4339 -96.8008 -81.4733 3 Conclusion Present investigation undertakes multiple linear regression (MLR) aided QSAR studies towards identification of structural insights of methoxy substituted chalcones. QSAR studies identified the profound linear relationship among structures of methoxy substituted chalcones and their biological activity measures IC50 (nM). It concludes that any structural variation in present class of chalcones would bring a linear change in IC50. Tetra-variable QSAR model was found statistically fit (R=0.809) and significant in predictive powers. We designed new candidates furnishing the knowledge of present QSAR studies. Descriptor based inverse QSAR approach has been applied to computational modeling of new small molecules as chalcones derivatives. Furthermore, they have been compared with present dataset of methoxy substituted chalcones (synthesized) using molecular docking studies. In course of molecular docking studies newly designed molecules yielded promising results in terms of candidate TC-14 showing better binding capacity (Rerank Score -103.089) than dataset used in supervised training of QSAR studies.References 1.Milne G.W.A., Pharmacophore and drug discovery, In Encyclopedia of computational chemistry; von Rague Schleyer, P., Ed.; Wiley: New York, USA, (1998)2.Dror O., Shulman-Peleg A., Nussinov R. and Wolfson H.J., Predicting molecular interactions insilico, February 6, (2004)3.Li X. and Lin J., QSAR modeling, J. Math. Chem., 33, 81–89 (2003)4.Vogel S., Barbic M., Juergenliemk G. and Heilmann J., Synthesis, cytotoxicity, anti-oxidative and anti-inflammatory activity of chalcones and influence of A-ring modifications on the pharmacological effect, Eur. J. Med. Chem., 45(6), 2206-2213 (2010)5.Lv P.C., Li D.D., Li Q.S., Lu X., Xiao Z.P. and Zhu H.L., Synthesis, molecular docking and evaluation of thiazolyl-pyrazoline derivatives as EGFR TK inhibitors and potential anticancer agents, Bioorg. Med. Chem. Lett., 21(18), 5374-5377 (2012)6.Mourad M.A.E., Abdel A.M, Abuo R.G.D. and Farag H.H., Design, synthesis and anticancer activity of nitric oxide donating/chalcone hybrids, Eur. J. Med. Chem., 54, 907-913 (2012)7.Mullen L.M.A., Duchowicz P.R. and Castro E.A., QSAR treatment on a new class of triphenylmethyl-containing compounds as potent anticancer agents, Chemometr. Intell. Lab., 107(2), 269-275 (2012)8.Sivakumar P.M., Iyer G., Natesan L. and Doble M., 3'-Hydroxy-4-methoxychalcone as a potential antibacterial coating on polymeric biomaterials, Appl . Surf. Sci., 256(20), 6018-6024 (2010)9.Bandgar B.P. and Gawande S.S., Synthesis and biological screening of a combinatorial library of [beta]-chlorovinyl chalcones as anticancer, anti-inflammatory and antimicrobial agents, Bioorg. Med. Chem., 18(5), 2060-2065 (2010) Research Journal of Chemical Sciences ___________________________________________________________ ISSN 2231-606XVol. 4(3), 76-80, March (2014) Res. J. Chem. Sci. International Science Congress Association 80 10.Lai C.H., Rao Y.K., Fang, S.H, Sing Y.T. and Tzeng Y.M., Identification of 3', 4', 5'-trimethoxychalcone analogues as potent inhibitors of Helicobacter pylori-induced inflammation in human gastric epithelial cells, Bioorg. Med. Chem. Lett., 15; 20(18), 5462-5465 (2010)11.Yang E.B., Guo Y.J., Zhang K., Chen Y.Z. and Mack P., Inhibition of epidermal growth factor receptor tyrosine kinase by chalcone derivatives, Biochim. Biophys. Acta., 1550(2), 144-152 (2001)12.Khatib S., Nerya O., Musa R., Shmuela M., Tamira S. and Vaya J., Chalcones as potent tyrosinase inhibitors: the importance of a 2, 4-substituted resorcinol moiety,Bioorg Med Chem., 13(2), 433-441 (2005)13.Lawrence N.J., Mc. Gown A. T., Ducki S. and Hadfield J.A., The interaction of chalcones with tubulin, Anti. Canc. Drug. Des., 15(2), 135-141 (2000)14.Zhang H., Liu J.J., Sun J., Yang X.H., Zhao T.T., Lu X., Gong H.B. and Zhu H.L., Design, synthesis and biological evaluation of novel chalcone derivatives as antitubulin agents, Bioorg. Med. Chem., 15;20(10), 3212-3218 (2012)15.Mojzis J., Varinska L., Mojzisova G., Kostova I. and Mirossay L., Antiangiogenic effects of flavonoids and chalcones, Pharmacol. Res., 57(4), 259-265 (2008)16.Pilatova M., Varinska L., Perjesi P., Sarissky M., Mirossay L., Solar P., Ostro A. and Mojzis J., In vitro antiproliferative and antiangiogenic effects of synthetic chalcone analogues, Toxicol. in vitro, 24(5), 1347-1355 (2010)17.Hijova E., Bioavailability of chalcones, Bratisl. Lek. Listy., 107(3), 80 (2006)18.Kalani K., Yadav D. K., Khan F., Srivastava S.K., Suri N., Pharmacophore, QSAR, and ADME based semisynthesis and in vitro evaluation of ursolic acid analogs for anticancer activity, J. Mol. Model., 18(7), 1-25 (2012)19.Liu T., Tang G.W. and Capriotti E., Comparative modeling: the state of the art and protein drug target structure prediction. Comb. Chem. High Throughput Screen, 14(6), 532-47 (2011) 20.Yang J.M. and Chen, C.C., GEMDOCK: A generic evolutionary method for molecular docking, Proteins, 55(2), 288-304 (2004) 21.Gehlhaar D.K., Verkhivker G., Rejto P.A., Fogel D.B., Fogel L.J. and Freer S.T., Docking conformationally flexible small molecules into a protein binding site through evolutionary programming, Proceedings of the Fourth International Conference on Evolutionary Programming, San Diego, Cambridge, 615-27 (1995)22.Gehlhaar D.K., Bouzida D. and Rejto P.A., Fully automated and rapid flexible docking of inhibitors covalently bound to serine proteases. Proceedings of the Seventh International Conference on Evolutionary Programming, San Diego, Cambridge, 449-461 (1996) 23.Pandey, N., Yadav, M, Nayarisseri, A., Ojha, M., Prajapati, J. Cross evaluation of different classes of alpha-adrenergic receptor antagonists to identify overlapping pharmacophoric requirements, J. Pharm. Research,6(1), 173-178 (2013)24.Bachwani M., Kumar R., Molecular Docking: A review,IJRAP, 2(6), 1746-1751 (2011)