Structurebased drug design and AutoDock study of potential protein tyrosine kinase inhibitors.

Different classes of compounds were investigated for their binding affinities into different protein tyrosine kinases (PTKs) employing a novel flexible ligand docking approach by using AutoDock 3.05 and 4. These compounds include many flavin analogs, which were developed in our group with varying degrees of cytotoxic activity (comparable or moderately superior to cisplatin and ara-c), and database selected analogs. They were docked onto twelve different families of PTKs retrieved from the Protein Data Bank. These proteins are representatives of plausible models of interactions with chemotherapeutic agents. A comparative study of the intact co-crystallized ligands of various types of PTKs was carried out. Results revealed that the new class of 5-deazapteridine and steroid hybrid compounds VIa,b, and d, and the vertical-type bispyridodipyrimidine with n-hexyl chain junction between its N-10 and N-10 atoms Xa, exhibited non-selective PTK binding capacities, with the lowest (Gb). On the other hand, 2-amino benzoic acid analog IIa, phenoxypyrido [3, 4-d]pyrimidine derivative IVc, tyrosine containing tripeptide Vd, and the one from Sumisho data base 831 are proposed to have selective PTK binding affinities to certain classes of tyrosine kinases, namely, HGFR (c-met), ZAP-70, insulin receptor kinase, EGFR, respectively. All These compounds of highest affinities were docked within the binding sites of PTKs with reasonable RMSD and 1-5 hydrogen bonds.


Background:
In 2009, two kinase inhibitors were approved for the treatment of cancer; and another for the treatment of renal cell carcinoma [1]. Protein-tyrosine kinases (PTKs) play an important role in the control of most fundamental cellular processes including the cell cycle, cell migration, cell metabolism, and cell survival, as well as cell proliferation and differentiation [2]. Docking of the molecules into their respective 3D macromolecular targets is a widely used method for lead optimization [3]. Docking programs find their most important applications in virtual database screening approaches in which hundreds of thousands of molecules are docked into the binding pocket to identify plausible binders [4]. It was reported that docking programs are able to predict experimental poses with deviation average from 1.5 to 2.0 Å root mean square deviation (RMSD) [5]. One of the most well-known docking programs is Autodock [6]. Structure-based drug design methods utilize knowledge of the three-dimensional structure of a receptor complexed with a lead molecule in an attempt to optimize the bound ligand or a series of congeneric molecules. Using a model with a given structure, a medicinal chemist can compute an activity of a molecule [7].

Development of PTK signaling inhibitors has evolved into an important approach toward new therapeutics [8].
Tyrosine kinase inhibitors as a target for anti-angiogenesis can be aptly applied as a new mode of cancer therapy [9]. The role of tyrosine kinases in cancer molecular pathogenesis is immense and recently tyrosine kinases have come in vogue as potential anticancer drug targets, sending a number of anticancer drugs to the market. Several tyrosine kinase inhibitors are undergoing human trials and several are in the pipeline of drug discovery [10].
The main objective of this study is to design and to propose potential tyrosine kinase inhibitors from synthesized compounds and databaseselected analogs. In this study twelve different target protein tyrosine kinases were selected to investigate the potential binding affinities within their corresponding binding sites. Those include fibroblast growth factor receptor, epidermal growth factor receptor, hepatocyte growth factor receptor, insulin receptor kinase, sarcoma proto-oncogenic kinase, spleen tyrosine kinase, platelet-derived growth factor receptor, vascular endothelial growth factor receptor, Abelson tyrosine kinase, hemopoietic cell kinase, Zeta-chain-associated protein kinase, and Janus kinase.

Methodology: AutoDock protocol:
Firstly, all bound waters, ligands and cofactors were removed from the proteins. The macromolecule was checked for polar hydrogens, partial atomic Kollman charges were assigned, and then atomic solvation parameters were allotted. Torsion bonds of the inhibitors were selected and defined. Secondly, the three dimensional grid box was created by AutoGrid algorithm to evaluate the binding energies on the macromolecule coordinates. The grid maps representing the intact ligand in the actual docking target site were calculated with AutoGrid (part of the AutoDock package). The three dimensional grid box with 60 Å grid size (x, y, z) with a spacing of 0.300 Å, grid was created. Eventually cubic grids encompassed the binding site where the intact ligand was embedded. Finally, AutoDock was used to calculate the binding free energy of a given inhibitor conformation in the macromolecular structure while the probable structure inaccuracies were ignored in the calculations.

Target macromolecules investigated:
Twelve different target protein tyrosine kinases were investigated, and those were retrieved from the Protein Data Bank, http://www.rcsb.org/pdb/home/home.do. For each docking target, crucial amino acids of the active site were identified using data in PDBsum, http://www.ebi.ac.uk/pdbsum/.

Analysis of the docking results:
Accelrys DS modeling 1.7 software [DS modeling 1.7; Accelrys Inc., San Diego, CA, (2006), www.accelrys.com], and Accelrys DS Visualizer 2.5 were utilized for molecular modeling, for evaluation of hydrogen bonds in ligand-receptor interaction, and for measuring RMSD which was computed and expressed in angstrom (Å) as a locational comparison of two molecules of interest. In our particular study RMSD was measured as distance between the centroids of the docked inhibitor and the intact ligand.

Molecular docking study:
Computer simulated automated docking studies were performed using the widely distributed molecular docking software, AutoDock versions 3.05 and 4.0, grid-based docking programs [6], which was utilized for the study of binding mode of the suggested potential inhibitors within different tyrosine kinases. In AutoDock, the overall docking energy of a given ligand molecule is expressed as the sum of intermolecular interaction energies including van der Waals attractive and repulsive energies, Hbond interaction energy, coulombic electrostatic energy, and the internal steric energy of the ligand [4]: Evaluation of docking performance and accuracy: As shown in Table 1 (see Supplementary material), the RMSD values for the representative examples range from 0.07-1 Å to 16 Å. It was found that the docked ligands are exactly superimposed with the originally embedded intact ligands not only for the best-scored conformations, but also for all the docked conformations. The docked compound (PRC1), 1terbutyl-3-p-tolyl-1h-pyrazolo[3, 4-d]pyrimidin-4-ylamine ligand, exhibited the best RMSD of 0.07Å, and a low binding free energy ( Gb) of -11.56 Kcal/mol. The representative examples of PTKs exhibited fairly reasonable docking results well comparable to their originally embedded intact ligands. Hence the obtained docking results were well correlated with the biological methods [20].

Docking evaluation for ABL inhibitors:
The activation of ABL tyrosine kinanse causes chronic myelogenous leukemia (CML). Therefore, inhibitors of this PTK is proposed to be effective in the treatment of CML. Thus herein compound VIIIb, N3phenyl-8-chloro-N10-(3, 4-xylyl)-5-deazaflavin, was tested for docking into Abelson tyrosine kinase (ABL, pdb code: 1fpu) to find potential inhibitor candidate. Compound VIIIb exerted binding free energy ( Gb) of -11.09 kcal/mol, with RMSD of 4.49Å, and one hydrogen bond between its 2-oxo group and OH of Thr315 was formed.

Docking evaluation for PDFGR inhibitors:
The activity of the c-Kit receptor protein-tyrosine kinase is tightly regulated in normal cells, whereas the deregulated c-Kit kinase activity is implicated in the pathogenesis of human cancers Based on its finding, compound VIIl, 2-deoxo-2-tyrosino-5-deazaflavin, was studied for its potential binding into platelet-derived growth factor receptor (PDGFR,ckit, pdb code: 1t46) as shown in Table 1. It was found that the docked compound VIIl was bound to Tyr670 identically to the intact ligand STI-571 by two hydrogen bonds via 2-tyrosinyl COOH group, exerting binding free energy (ΔGb) of within -11.32 kcal/mol.

Docking evaluation for ZAP-70 inhibitors:
The ZAP-70 tyrosine kinase plays a critical role in T cell activation as well as in the immune response, and thus it has become a logical target for immunomodulatory therapies. Based on this background, against ZAP-70 tyrosine kinase (pdb code: 1u59) compound IIb, 4-(3, 5-di-tertbutylphenylcarbamoyl)-2-nitro benzoic acid, was docked within the binding site resulting in a strong binding affinity.

Docking evaluation for insulin receptor kinase inhibitors:
Tyrosine containing tripeptide, Trp-Glu-Tyr-OH Vd is structurally similar to the structure of the intact ligand of the insulin receptor kinase (pdb code: 1rqq), thiophosphoric acid o-(adenosyl-phospho)phospho)-sacetamidyl-diester (112 E), where there is a ring equivalence between the indole and adenine rings of both compounds. In addition, the structural similarity exists between of tripeptides of Vd and triphosphate moiety of 112E. The docked compound Vd was exactly superimposed on and identically to the intact ligand within 1.32Å RMSD exhibiting five hydrogen bonds between the pair of its COOH groups, indolyl-NH, and phenoxy-p-OH and amino acids of the binding sites, Gln1004, Lys1030, Met1079, and Ser1006, respectively, as shown in Table 1(see Supplementary material).

Docking evaluation for HCK inhibitors:
The docked compound VIIIg, was shown to be superimposed with the intact ligand PP1-532 of non-receptor tyrosine kinase, (HCK; pdb code: 1qcf) and was bound at the ATP-binding site of the kinase, exhibiting one hydrogen bond with lys295. Its lipophilic n-dodecyl moiety comparable to the p-tolyl moiety of intact ligand PP1-532 was inserted into an adjacent hydrophobic pocket created by VaL281, Lys295, Ile336, and Asp404.
Replacement of pyrimidoquinoline ring of 5-deazaflavins or benzopteridine ring of flavins with pyridodipyrimidines (IXc) as a new NAD-type redox catalyst produced compounds with excellent docking binding affinities wherein 2,7,8,10-tetraphenyl substitution enabled planar interaction within the target site so as to produce favorable docking affinity. Moreover, tyrosine containing tripeptide, Trp-Glu-Tyr-OH (Vd), was found to compose a platform to produce small lead compounds as PTK inhibitors which exhibit selective insulin receptor kinase binding affinity due to its close structural resemblance to the intact ligand of the insulin receptor kinase (pdb code: 1rqq).
The amino acid coupled with 2-deoxo-5-deazaflavin and flavins (VII l, q, r) has been shown to enhance remarkably the binding affinities in PTK. The higher affinity of these designed derivatives is presumably attributed to the formation of more hydrogen bonds (2-4 hydrogen bonds) between their 2-amino acid moieties and the binding site residues.