QSAR AND PHARMACOPHORE MODELING BASED DRUG DESIGNING FOR SPLEEN TYROSINE KINASE (SYK) PROTEIN FOR HUMAN USING ACCELRYS DISCOVERY STUDIO SOFTWARE IN LINUX SERVER

QSAR AND PHARMACOPHORE MODELING BASED DRUG DESIGNING FOR SPLEEN TYROSINE KINASE (SYK) PROTEIN FOR HUMAN USING ACCELRYS DISCOVERY STUDIO SOFTWARE IN LINUX SERVER

Kuldeep Sahu

Biocampus, GVK Biosciences Pvt. Ltd., Plot no-79, IDA Mallapur, Hyderabad-500 076, Andhra Pradesh, India

ABSTRACT: From this current research, Syk (spleen tyrosine kinase) protein and gene information is analyzed by different genomics, proteomics tools & databases. One crystal ligand 4DFL was collected from protein data bank (pdb). From different literature review 131 syk protein inhibitors were collected. Molecular modeling of these 131 molecules was done through Accelrys discovery studio (ADS). Choose appropriate force-field & minimization (Smart, Stephent Descent, and Conjugate Gradient) according to selected molecules. Then collected crystal ligand is purified by protein purification method and used appropriate conformation (BEST, FAST, and CAESAR). Docking methods were analyzed with protein, crystal ligand and similar inhibitors to know the best protein-ligand interaction. Pharmacophore research is done through HIPHOP and HYPOGEN method. Protein with final compound docking method is done after completion of virtual screening method.  Pharmacophore research with final molecule was done. Quantitative structure activity relationship (qsar) method is analyzed to know the correlation between the above selective structures. From virtual screening method, best and final compound is analyzed. So, final molecule can be a drug molecule for SYK protein abnormality diseases. However, the scope for fine tuning and optimizing this potent class of syK inhibitors could lead to the generation of new therapeutic agents.

Accelrys discovery studio, SYK Protein, Virtual screening, QSAR, Pharmacophore, Docking

INTRODUCTION:Spleen Tyrosine Kinase (syk) is best known as a non-receptor tyrosine kinase. It mediates signal transduction downstream of a variety of transmembrane receptors including classical immune receptor like B-cell receptor. It is involved B-cell and T-cell function, regulates several biological processes including innate & adaptive immunity and plays also a crucial role in the innate immune response to fungal, bacterial and viral pathogens.

It is also involved in cell adhesion, mast cell signaling, osteoclast maturation, platelet activation and vascular development, neutrophils and macrophages. It is for instance involved in vascular development where it may regulate blood and lymphatic vascular separation, assembles into signaling complexes with activated receptor at the plasma membrane via interaction between sh2 domains and the receptor tyrosine phosphorylated ITAM domains.

Abnormality of this protein may cause the several diseases like Rheumatoid Arthritis, Naso-pharyngeal Carcinoma, all type of blood cancer (WBC, RBC and platelet), head cancer, neck cancer, breast cancer, gastric cancer and several autoimmune diseases ^1-9.

Drug designing is the inventive process of finding new medications based on the knowledge of a biological target. Computer-aided drug design uses computational chemistry to discover, enhance, or study drugs and related biologically active molecules. Structure-based drug design (or direct drug design) relies on knowledge of the three dimensional structure of the biological target obtained through methods such as x-ray crystallography or NMR spectroscopy ^10-14.

MATERIALS & METHODS:

Linux Operating System: (centOS): Operating system is an interface between hardware and application software. Feature of this operating system are multi-user, multi-tasking, portability, machine-independence, well security & 10 times faster than ordinary operating systems.

Accelrys Discovery Studio (2.5): Discovery studio is a software suite of life sciences molecular design solutions for computational chemists and computational biologists. It makes easier to examine the properties of large and small molecules, study systems, identify leads and optimize candidates.

Furthermore, built on Accelrys pipeline pilot technology, discovery studio enables scientists to rapidly automate routine tasks, integrate third party applications and even deploy models out to research colleagues together, this uniquely  positions discovery studio as a truly as a comprehensive collaborate research  solutions for both experts and project terms alike.

NCBI: http://www.ncbi.nlm.nih.gov/: The National Center for Biotechnology Information (NCBI) provides a comprehensive website for biologists that includes biology related databases, and tools for viewing and analyzing the data inherent in the databases. NCBI is an agency responsible for creating automated systems for storing and analyzing the rapidly growing profusion of genetic and molecular data.

PDB: http://rcsb.org/pdb/home/: The protein data bank (pdb) was established at Brookhaven national laboratories (BNL) in 1971 as archive for biological macromolecular crystal structure it is the single archive of biological macromolecular structures. The research Collaborator for Structural Bioinformatics (RCSB) has been fully responsible for its management since July 1, 1999.

Molecular Modeling: It is a collective term that refers to theoretical methods and computational techniques to model or mimic the behavior of molecules. The techniques are used in the fields of computational chemistry, computational biology ^{15, 16}.

Force field: A force field is used to calculate the energy and geometry of a molecule. It is collection of atom types (to define the atoms in a molecule), parameters (for bond, lengths, bond angles, etc) and equations (to calculate the energy of a molecule). a set of functions and parameterization used in molecular mechanics calculations, 1^st generation Force Field (FF) applicable mainly to biochemistry. 2^nd generation FF capable of predicting many properties, rule based FF applicable to a broad range of periodic table. Special purpose of FF narrowly applicable to particular applications or types of models ^{17, 18}

Minimization: A process by which a molecular structure is brought to minimum energy conformation. It involves iteratively adjusting atomic coordinates until the forces acting on the atoms are zero or close to zero. Minimization generally takes the molecule to the local minimum nearest to the starting conformation ^{17, 18}.

Docking: Docking is an energy optimization process concerned with the search of the lowest free energy binding mode of a ligand within a protein binding site. Docking is a method which predicts the preferred orientation of one molecule to a second when bound to each other to form a stable complex. It has 2 components (pose search, scoring).

Pharmacophore: Pharmacophore is an analog based method. This word is coined by Paul Ehrlich in the early 1900s referring to a molecular frame work that carries the essential features (phoros) responsible for a drug’s (pharmacon) biological activity. In 1977 Peter Gund redefined as “a set of structural features in a molecule that is recognized at a receptor site and is responsible for that molecule’s biological activity”. A pharmacophore can also be generated from the receptor structures.

In dynamic pharmacophore model based on molecular dynamics trajectories takes care of the binding site dynamics. These models can be used for optimizing known ligands or for screening databases to find potential novel leads suitable for further development ¹⁹.

Virtual screening: Virtual screening as "automatically evaluating very large libraries of compounds" using computer programs. The aim of virtual screening is to identify molecules of novel chemical structure that bind to the macromolecular target of interest. Virtual screening is very important part of drug design and drug discovery research.^20-23.

Quantitative Structure Activity Relationship (QSAR): Quantitative structure activity relationships (qsar) represent an attempt to correlate structural or property descriptors of compounds with activities. Sufficient number of ligands active against target of intrest should be available to develop the structure activity relationships. The equations that are parameterized for one target do not apply to another.

A TOTAL OVERVIEW OF THIS RESEARCH

fig. 1a: molecule shows the best interactions with important amino acids with libdock Docking method

Fig-2: hiphop pharmacophore best fit with best molecule obtained from hiphop pharmacophore result

TabLE 1: hiphop exporting hypotheses from hiphop algorithm

fig. 3: hypogen pharmacophore best fit with best molecule obtained from hypogen pharmacophore result

TabLE 2: hypogen exporting hypotheses from hypogen algorithm

tabLE 3: It shows the best hypothesis of hypogen training set

Fig. 4: pharmacophore hypogen test graph

Fig. 5(a) shows the Best compound after screening, Fig. 5(b) shows best compound with best interaction with important amino acids with LIBDOCK docking method

Fig. 6: Best drug molecule with best fit result in pharmacophore

Fig-7: QSAR train set graph obtained from ADS software

Fig-8: qsar test set analysis graph

discussions: Spleen Tyrosine Kinase (syk) is a protein best known as non-receptor tyrosine kinase. Abnormality of this protein may cause various diseases like Rheumatoid Arthritis, Nasopharyngeal Carcinoma, and all type of blood cancer, head cancer, neck cancer, breast cancer, gastric cancer and several autoimmune diseases. Gene and protein information are analyzed by different online tools, databases (NCBI, KEGG, GeneCards, pdb, Uniprot) and literature review.

One crystal ligand (4DFL - which encodes for syk protein) is collected from pdb database on the basis of <3.0 resolution, latest and updated, less sequence gap in protein active site, which is mentioned in run ligplot pdf (downloaded from PDB).

From literature review, 131 syk protein inhibitors were collected. Molecular modeling of these inhibitors is done by Accelrys Discovery Studio. Force fields are 2 types (fitted force fields, rule based).We have to research which is an appropriate Forcefield (charm, CHARMm polar H, CVFF, charmm19, charmm22, charmm27, cff, MM3, xprolig, mmff, ESFF, UFF) with respective partial charges for our selective inhibitor molecules. We have to analyze which is the appropriate minimization (smart, steepest-descents, and conjugate-gradients, Powel) and adjusting the steps for our selective inhibitors.

Potential and absolute energy were noted. This above method is called molecular modeling. Then protein is purified (deletion of unnecessary ligands, water molecules, clean protein and so on) by protein purification method. Steepest descent and conjugate gradient minimization is applied one by one and results are noted respectively. We have to analyze which is the most preferable docking method (ligand fit, cdocker, libdock, and DENOVO) for our selective protein.

From 131 molecules, 18 molecules (low, moderate, high energy molecules) were selected and to get an overview which is the best docking method for our selective protein inhibitors. The Results concluded that libdock docking method is most preferable for syk protein. In this method high energy molecules are interacting with all important amino acids (glu 449, ALA 451 and ASN 499) with distance 2.174, 1.8750, 2.92881 respectively shown in fig. 1.

In pharmacophore, hiphop or common feature pharmacophore method with 6 high energy molecules were selected from 131 syk inhibitors. At a time 6 molecules were opened in ADS software.  2 attributes (principal, MaxOmitFeat) were added. The features of these attributes are (2-0, 2-0, 1-1, 1-1, 0-2 and 0-2). The minimum interfeature distance of this hiphop algorithm is 2.97 with best conformation. The features are (hydrogen bond acceptor (A), hydrogen bond donor (D), hydrophobic (H), hydrophobic aromatic (Y), hydrophobic aliphatic (Z)).which shown in fig. 2. hiphop exporting hypotheses from hiphop algorithm is shown in Table 1.

In hypogen method, the total syk inhibitors are divided in two parts i.e. training set and test set. Two attributes (Active, Uncert) are added in training set and test set, the features of these 2 attributes are (active value, Uncert is 3). In training set 24 low, moderate and high active molecules were selected.pharmacophore HYPOGEN algorithm with best fit molecule is shown in fig. 3. Results obtained from 10 hypotheses of hypogen algorithm. Total cost, cost difference, error cost, RMS, correlation, feature (adhr) are noted, shown in table 2.

Out of these 10 hypothesis was considered as the best due to high cost difference from that null hypothesis. From our best result where total cost is 101.894, cost difference is 43.816, error cost is 90.9679, RMS is 0.923821, and correlation is 0.918158. Cost difference is calculated by this formula (null cost-total cost), shown in table 3.  Another 107 molecules are taken as test set. From this 107 molecules pharmacophore hypogen test graph was obtained. Further validation of the hypothesis was done using the test set molecules for ligand pharmacophore mapping.

A graph was plotted between activity and estimated activity, where R² value is 0.723 (according to our rule R²value must be equal or greater than 0.7), shown in fig. 4. In virtual screening/library screening method all molecules were selected. In that database ADS will create 2000 similar molecules as per our selected SYK inhibitors. The features of virtual screening are input ligand is all ligand, input type is database, minimum features is 4, maximum features is 5, maximum subset of pharmacophore is 100, minimum inter feature distance 2.0, fitting method is rigid and with best conformation.

From that database, it will find 4-5 best compounds. From these compounds, we have to analyze which is best molecule for abnormality of syk protein. Results concluded that BTB06517 is the best compound, shown in fig. 5(a) where absolute energy is 55.2114; estimate value is 25348.5; best fit value is 3.81925; relative energy is 4.5031. In libdock docking it is interacting with important amino acids asp 499, ala 451 with distance 2.51639, 2.98385 respectively, shown in fig. 5(b).

Pharmacophore research is analyzed with this drug molecule which shown in fig. 6. The 3D QSAR research conducted for training set gave good R² values of 0.729 with MLR graph with a fit that representing the good correlation of the compounds with the activities. QSAR 2D and 3D were analyzed with our selected total molecules. In qsar we have divided into training and test set with 2:1(85:46) ratio.

The molecular properties are AlogP, molecular_weight, Num_aromatic rings, Num_H_Acceptors, Num_H_donor, Num_rings, Num_rotatable bonds, molecular functional polar surface area, Dipole_mag, Dipole_X, Dipole_Y, Dipole_Z, Jurs_DPSA_1, Jurs_DPSA_2, Jurs_DPSA_3, Jurs_FNSA_1, Jurs_FNSA_2, Jurs_FNSA_3, Jurs_FPSA_1, Jurs_FPSA_2, Jurs_FPSA_3, Jurs_PNSA_1, Jurs_pnsa_2, Jurs_pnsa_3, Jurs_PPSA_1, Jurs_ppsa_2, Jurs_PPSA_3, Jurs_RASA, Jurs_rncg, Jurs_RPCS, Jurs_RPSA, Jurs_SASA. After calculations of molecular properties one PIC50 attribute is added. PIC50 value is calculated by this formula “9-log (active value of molecule)”.

In training set molecules were taken as the 2D molecular properties were calculated. The multilinear regression algorithm was applied and a graph was plotted between activity and the QSAR_mlr model obtained, where R² value is 0.729, shown in fig. 7. Result of all 46 molecules taken as test set. With the help of train set result and with same procedure followed for test set. QSAR test set graph is shown in fig. 8.

conclusion: In silico drug design studies considered for syk protein. The algorithms such as MOLECULAR MODELING, PROTEIN PURIFICATION, docking, pharmaco-phore, virtual screening and qsar were used. These algorithms showed good results and further investigation for the drug collaboration can be done. Results concluded that btb06517 compound having good interactions with docking method.

It had a good potential, best fit with other molecules and best fit with our pharmacophore result. So this compound may be a drug molecule for syk (Spleen Tyrosine Kinase) abnormality diseases. Lead optimization and lead to the generation of a highly potent series of SYK inhibitors with good drug like properties. However, the scope for fine tuning and optimizing this potent class of syK inhibitors could lead to the generation of new therapeutic agents.

ACKNOWLEDGEMENT: Author would like to extend sincere thanks to Dr. J. A. R.P Sharma, Sr. Vice President, Dr.K.V.Radha kishan, Director, Dr. Rambabu Gundla, Principal and Mr. Sudheer Kumar Ready, training manager for giving a chance to carry out a research work at bio-campus. Author is also thankful to his guides Mr. Manohar Suram, Mrs. Umadevi Gajjala and Mr. Sardar Shamshair Singh of Biocampus, GVK Biosciences Pvt. Ltd. who helped and gave a proper guide to learn protein modeling and rational drug designing using Accelrys Discovery Studio in Linux server.

references:

1. Law CL, Chandran KA, Sidorenko SP, Clark EA.:Phospholipase C-gamma1 interacts with conserved phosphotyrosyl residues in the linker region of Syk and is a substrate for Syk: Molecular Cell Biology 1996; 16(4):1305-15.
2. Deckert M., Elly C., Altman A., Liu Y.C.: Coordinated regulation of the tyrosine phosphorylation of Cbl by Fyn and Syk tyrosine kinases: Journal of Biological Chemistry 1988; 273:8867-8874.
3. Lupher M.L. Jr., Rao N., Lill N.L., Andoniou C.E., Miyake S., Clark E.A., Druker B., Band H:Cbl-mediated negative regulation of the Syk tyrosine kinase. A critical role for Cbl phosphotyrosine-binding domain binding to Syk phosphotyrosine 323: Journal of. Biological chemistry 1988; 273:35273-35281.

4. Chiu C.W., Dalton M., Ishiai M., Kurosaki T., Chan A.C.: BLNK: molecular scaffolding through 'cis'-mediated organization of signaling proteins: EMBO Journal 2002; 21:6461-6472.
5. Tollenaere JP., the role of structure-based ligand design and molecular modelling in drug discovery: Pharmacy and world sciences 1996; 18(2): 56-62.
6. Renxiao Wang, Ying Gao and Luhua Lai ; Lig Builder: A Multi-Purpose Program for Structure-Based Drug Design, Journal of Molecular Modeling 2000; 6: 498-516.
7. Jorgensen WL.: The many roles of computation in drug discovery: Science 2004; 303: (5665).
8. F. Ooms: Molecular Modeling and Computer Aided Drug Design Examples of their Applications in Medicinal Chemistry: Current Medicinal Chemistry 2000; 7:141-158.
9. Facultés F. Universitaire, Notre-Dame de la Paix: Molecular Modeling and Computer Aided Drug Design. Examples of their Applications in Medicinal Chemistry: Current Medicinal Chemistry 2000; 7:141-158.
10. Michael Crowley, Tom Darden, Thomas Cheatham and David Deerfield: Adventures in improving the scaling and accuracy of a parallel molecular dynamics program: The Journal of Supercomputing 1997; 11, 225-278.
11. Wilfred Vangunsteren and Herman Berendsen: Computer simulation of molecular dynamics; methodology, applications, and perspectives in chemistry: Angewandte Chemie International Edition1990; 29: 992-1022.
12. Bernard brooks, Robert Bruccoleri, Barryolafson, David States, Swaminathan and Martin Karplus: CHARMM: a program for macromolecular energy, minimization and dynamics calculation: Journal of computational chemistry1982; 4(4):187-217.
13. James Vincent and Kenneth Merz: A highly portable parallel implementation of AMBER4 using the message passing interface standard: journal of computational chemistry 1995; 16(11):1420-1427.
14. Kaminski. J.J, Rane D. f, Rothossky, M.L: Pharmacophore perception, development and use in drug design, Osman F. Gilner, ed.; International university line: La Jolla, California; chapter 14, 2000: 251-268.
15. Rester U.: From virtuality to reality - Virtual screening in lead discovery and lead optimization: a medicinal chemistry perspective: Current Opinion in Drug Discovery and Development 2008; 11(4): 559-568.
16. Patrick Walters W., Matthew T Stahl, Mark A Murcko: Virtual screening-an overview: Drug Discovery Today 1998; 3: 160-178.
17. Srinivas Reddy A, Priyadarshini Pati S, Praveen Kumar P, Pradeep HN and Narahari G Sastry: Virtual Screening in Drug Discovery – A Computational Perspective: Current Protein and Peptide Science 2007; 8: 329-351.
18. Urzainqui A., Serrador J.M., Viedma F., Yanez-Mo M., Rodriguez A., Corbi A.L., Alonso-Lebrero J.L., Luque A., Deckert M., Vazquez J.,Sanchez-Madrid F.: ITAM-based interaction of ERM proteins with Syk mediates signaling by the leukocyte adhesion receptor PSGL-1:Immunity 2002 17:401-412.
19.  Shim E.K., Moon C.S., Lee G.Y., Ha Y.J., Chae S.K., Lee J.R.:Association of the Src homology 2 domain-containing leukocyte phosphoprotein of 76 kD (SLP-76) with the p85 subunit of phosphoinositide 3-kinase: FEBS Letter 2004; 575:35-40
20. Cholay M., Reverdy C., Benarous R., Colland F., Daviet L.: Functional interaction between the ubiquitin-specific protease 25 and the SYK tyrosine kinase: Experimental. Cell Research 2010; 316:667-675.
21. Bohmer R, Neuhaus B, Bühren S, Zhang D, Stehling S, Böck B and Friedemann Kiefer . Regulation of Developmental Lymph angiogenesis by Syk⁺ Leukocytes: Developmental Cell 2010; 18 (3): 437–49.
22. Weinblatt ME, Kavanaugh A, Genovese MC, Musser TK, Grossbard EB, Magilavy DB: An oral spleen tyrosine kinase (Syk) inhibitor for rheumatoid arthritis: The New England Journal of Medicine 2010; 363 (14): 1303–12.
23. TingjunHou and Xiaojie Xu.: Recent Development and Application of Virtual Screening in Drug Discovery: Current Pharmaceutical Design 2004; 10: 1011-1033

How to cite this article:

Sahu K: Designing for Spleen Tyrosine Kinase (syk) protein for human using Accelrys discovery studio software in Linux server. Int J Pharm Sci Res 2013; 4(11): 4272-80. doi: 10.13040/IJPSR. 0975-8232.4(11).4272-80

All © 2013 are reserved by International Journal of Pharmaceutical Sciences and Research. This Journal licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License.