In silico studies on marine actinomycetes as potential inhibitors for Glioblastoma multiforme.

Glioblastoma multiforme (GBM) is considered to be the most common and often deadly disorder which affects the brain. It is caused by the over expression of proteins such as ephrin type-A receptor 2 (EphA2), epidermal growth factor receptor (EGFR) and EGFRvIII. These 3 proteins are considered to be the potential therapeutic targets for GBM. Among these, EphA2 is reported to be over-expressed in ˜90% of GBM. Herein we selected 35 compounds from marine actinomycetes, 5 in vitro and in vivo studied drug candidates and 4 commercially available drugs for GBM which were identified from literature and analysed by using comparative docking studies. Based on the glide scores and other in silico parameters available in Schrödinger, two selected marine actinomycetes compounds which include Tetracenomycin D and Chartreusin exhibited better binding energy among all the compounds studied in comparative docking. In this study we have demonstrated the inhibition of the 3 selected targets by the two bioactive compounds from marine actinomycetes through in-silico docking studies. Furthermore molecular dynamics simulation were also been performed to check the stability and the amino acids interacted with the 3 molecular targets (EphA2 receptor, EGFR, EGFRvIII) for GBM. Our results suggest that Tetracinomycin D and Chartreusin are the novel and potential inhibitor for the treatment of GBM.


Background:
Glioblastoma multiforme (GBM) is the most common and aggressive type of primary brain tumor in humans [1]. GBM is a fast-growing type of central nervous system (CNS) tumor arises mainly from glial tissue of the brain and spinal cord. GBM usually occurs in adults and affects the brain more often than the spinal cord [2]. It has been proved that 17,000 primary brain tumors diagnosed in the United States each year, out of which approximately 60% are gliomas [3]. Gliomas are a group of central nervous system (CNS) neoplasms with various histological characteristics. They are classified into two major groups as astrocytomas and oligodendrogliomas based on their morphological and histological resemblances between malignant and normal cells [4]. The most common form of gliomas in human is the astrocytoma, and the most aggressive type is GBM [5].
Molecular markers that are found on tumour cells and are also over-expressed on malignant cells were nearly absent on normal cells which facilitates as attractive drug targets. Along these lines we found ephrin-A2 (EphA2), epidermal growth factor receptor (EGFR) and EGFRvIII are considered to be novel molecular targets for this pathological condition, because it is expressed in high quantities in GBM [6]. EphA2 is a type of tyrosine kinase family [3] and this receptor is over expressed in various cancers of brain, breast, cervix, colon, oesophagus, head, neck, liver, lungs, ovary and skin [7][8][9][10][11][12][13][14][15][16]. The EphA2 is believed to be an ultimate target in many cancers [17]. It plays a critical role in embryonic patterning, neuronal targeting, and vascular development during normal embryogenesis, cell proliferation and migration [18]. Eph receptor tyrosine kinases (Eph RTKs) and their ligands, ephrins are frequently over expressed in a various types of cancer and tumor [19]. About fourteen Eph receptor and eight ephrin ligands are involved in the various type of cancer [20][21]. In vitro study has shown that modifications or a mutation in the EphA2 is responsible for GBM [22].
EGFR is also a member of tyrosine kinase family containing proteins and it plays a major role in signal transduction pathway responsible for cell differentiation and proliferation [23]. Over expression of EGFR in GBM has been proven in many in vitro and in vivo studies [24]. It is also been over expressed in 50-60% of gliomas [25]. EGFRvIII is over expressed in most 24-67% of glioblastoma patients [25]. This is a mutated type of the tyrosine kinase receptor. This is caused by genetic loss of 270 amino acids from the EGFR. This mutated type protein is responsible for GBM disease in humans [24]. Multimodality therapy for this disease still remains unsatisfied [26]. In this study we adopted cheminformatics-based drug design approach to identify potential inhibitors against GBM. We conducted comparative docking studies using molecular modelling approach against a total of 44 drug-like molecules which includes (i) 35

Docking using Glide extra precision:
All the ligands which were prepared using LigPrep were then subjected for docking against the 3 molecular targets including EphA2, EGFR and EGFRvIII (PDB ID: 1MQB, 1M17 and 1I8I) using Glide extra-precision (XP), version 5.5 [38] mode. The grid-enclosing box was centered to the active sites of the corresponding 3D-structures of these 3 molecular targets to GBM; so as to enclose them within 3 Å from the centroid of these residues. A scaling factor of 1.0 was set to van der Waals (VDW) radii for these residue atoms, with the partial atomic charge less than 0.25. Glide XP mode determines all reasonable conformations for each low-energy conformer in the designated binding site. In the process, torsional degrees of each ligand are relaxed, though the protein conformation is fixed. During the docking process, the glide scoring function (G-score) was used to select the best conformation for each ligand. Final Gscores were analysed based on the conformation at which the ligands formed hydrogen bonds to at least one of the active site amino acid residues of the corresponding 3D-structures of these 3 molecular targets with optimal binding affinity. Herein, the data obtained from these dockings were used to analyse the molecular interactions and also to identify the residues involved in hydrogen bond formation with 1MQB, 1M17, and 1I8I. The glide scores and energies including van der Waals (VDW) and electrostatic were calculated for all the ligands against EphA2, EGFR and EGFRvIII. Finally the molecular interactions and functional role of the two selected marine actinomycetes compounds named and the commercially available drugs were proposed in detail. All these docking procedures were performed on a Dell RHEL 5.0 workstation.

Molecular Dynamics Simulation:
Molecular Dynamics simulations were done with all the 3 molecular targets of GBM against the two selected bioactive compounds. The MD simulation was performed by using Gromacs 3.3.2 [39,40]. It works based on the leap-frog algorithm to integrate Newton equations. The NPT ensemble and Gromos96 force field were applied to the system. Each docking complex was placed in the center of a 72 Å × 72 Å × 72 Å cubic box and solvated by simple point charge water molecules (SPC/E). Na + counter ions were added to keep the system electrically neutral and the periodic boundary condition was also applied to the system. Energy minimization was carried out by using steepest-descent method [41]. Berendsen temperature and pressure coupling methods were applied to keep the system in stable environment (300 K, 1 Bar), and the coupling constants were set at 0.1 and 1.0 for temperature and pressure respectively. Cut-off method was employed for electrostatic and van der walls interactions; cut-off distance for the short-range neighbor list (rlist) and was set at 0.8, whereas coulomb cut-off (rcoulomb) and VDW cut-off (rvdw) was fixed at 1.4. The LINCS algorithm was used to constraint the bonds [42]. The simulation was performed with a time step of 2fs and the coordinates were saved every 1000 steps. 20 ps position restraining dynamics simulation was carried to relieve close contacts and to equilibrate the protein in the medium; finally 1ns molecular dynamics simulation and further analysis were performed. The dynamics results were visualized using VMD [43].

Assessment of drug-like properties of selected optimized ligands:
The selected optimized molecules were studied for their drug-like properties based on Lipinski parameters using QikProp version 3.2 [38], and also the percentage of their human oral absorption was also predicted to determine the toxicity levels, by use of QikProp

Results and Discussion:
In this study, we conducted a comparative docking and molecular dynamics simulation between the two selected bioactive molecules that include Tetracenomycin D and Chartreusin which was obtained from 35 marine actinomycetes compounds along with the 5 compounds derived from the experimental studies and 4 commercially available drugs against GBM. The in silico results revealed that the two bioactive molecules exhibited better binding affinity than the commercially available drugs against the 3 molecular targets of GBM including EphA2, EGFR and EGFRvIII.

Binding site analysis for the molecular targets against GBM:
The binding site for the two molecular targets including EphA2 and EGFR (PDB ID: 1MQB, 1M17) are known and were determined using PDBsum. But EGFRvIII (PDB ID: 1I8I) does not have any defined active site and hence it was predicted using SiteMap program in Schrödinger. The predicted amino acids were identified to be Asp408, Gln412, Trp410, Phe46, Gln301, Val302, and Gln412. Docking studies were performed with the two bioactive molecules against the 3 molecular targets based on their corresponding co-crystallized ligands available in their 3D-structures. All binding pockets of protein-ligand complexes were shown in the  The amino acids Met695 and Thr692 of PDB ID: 1MQB were identified to form hydrogen bonds with its co-crystallized ligand Phosphoaminophosphonic acid-adenylate ester present in the 3D-structure of this protein. Whereas the docking data after simulation with Tetracenomycin D and Chartreusin revealed that the amino acids Met695, Asn744 and Arg743 formed hydrogen bonds with EphA2. The amino acids Asn744 and Arg743 are the active site residues of 1MQB. Hence it has been observed that the binding of the two active bio molecules against EphA2 can block the active site region of this protein.
Whereas for EGFR (PDB ID: 1M17) the amino acid Met769 is involved in hydrogen bonding with the co-crystallized ligand 6, 7-bis(2-methoxyethoxy)quinazoline-4-yl]-(3-ethynylphenyl)amine [Erlotinib]. The residues interacted with EGFR with the two bioactive molecules were found to be Met769 which is found to interact with the co-crystallized ligand of 1M17. On the other hand for EGFRvIII (PDB ID: 1I8I) the amino acids interacted and formed hydrogen bonds after simulation are Asp408, Phe46, Val302, Gln301, Lys303, Gln412 and Trp410 out of which none of them are found to interact with the co-crystallized peptide of 1I8I. However these amino acids are found to be in close contact with EGFRvIII. Finally the simulation results revealed that the two bioactive molecules can efficiently block EphA2 than when compared to EGFR and EGFRvIII without much conformational change in the active site after 1ns simulation (Table 3 see Supplementary material). The temperature and pressure does not imply any changes in the conformation of the structure. The hydrogen bonds that have been formed between the protein and ligand after simulation are mostly present in the β-sheets and loop regions of the protein which contains the active site region for the catalysis of the substrate binding. Since they form the hydrogen bonds with by blocking the active site region of the protein, the docking results suggests that the two bioactive molecules could efficiently inhibit the functional activity of the target proteins of GBM.

ADME or pharmacokinetics prediction of the ligands:
Predict pharmacokinetic properties using the QikProp module of the Schrödinger 2009 software. QikProp settings determine which molecules are flagged as being dissimilar to other 95% of the known drugs. Predicted significant ADMET properties such as permeability through MDCK cells (QPlogMDCK), QikProp predicted log IC50 value for blockage of K + channels (QPlogHERG), QikProp predicted gut-blood barrier (QPPCaco) and violations of the Lipinski's rule of five (LROF) were reported in Table 4 (see Supplementary material). The number of stars indicates the deviations from the 95% of the known drugs. Percent of Human Oral absorption is based on number of metabolites, number of rotatable bonds, logP, solubility and cell permeability.
In accordance with Lipinski's rule of five, QikProp was used to evaluate the drug-likeness of the lead molecules by assessing their physicochemical properties. Their molecular weights were < 500 Daltons with < 5 hydrogen bond donors, < 10 hydrogen bond acceptors and a log p of < 5 (Table 4 see Supplementary material); these properties are well within the acceptable range of the Lipinski rule for drug-like molecules. These compounds were further evaluated for their drug-like behavior through analysis of pharmacokinetic parameters required for absorption, distribution, metabolism, excretion and toxicity (ADMET) by use of QikProp. For the two bioactive compounds, the partition coefficient (QPlogPo/w) and water solubility (QPlogS), critical for estimation of absorption and distribution of drugs within the body, ranged between ~ 0.7 and ~ 2043, cell permeability (QPPCaco), a key factor governing drug metabolism and its access to biological membranes, ranged from 0.004 to 2050, while the bioavailability and toxicity were from ~ 3.4 to ~ 0.4. Overall, the percentage human oral absorption for the compounds ranged from ~ 36 to ~ 79%. All these pharmacokinetic parameters are within the acceptable range defined for human use (Table 4 see Supplementary material), thereby indicating the selected two drug-like compounds their potential as drug-like molecules could be a potential inhibitor of therapeutic targets of GBM disease and further analysis can be performed through various experimental studies. Among various commercially available drugs against GBM, the best two bio active compounds from marine actinomycetes have good interactions with the GBM targets. Tetracenomycin D and Chartreusin have also been confirmed in both in vitro and in vivo studies for the different cancer treatments [27]. ADMET properties of these two compounds are under acceptable range. So, these drugs can be a potential inhibitor of therapeutic targets of GBM disease and further analysis can be performed through various experimental studies.

Conclusion:
In the present study we have performed comparative docking analysis of various compounds using Glide and the results are interpreted. EphA2 was identified as good target for GBM. For the first time we proposed in silico study to identify the potential small molecule inhibitor for EphA2, EGFR, EGFRvIII proteins against GBM.    1787.647 -3.339 3282.329 3 0 a Ligand IDs of pubchem database 5281394-Chartreusin, 196730-Tetracenomycin D, 5329012-Sunitinib, 5426-Thalidomide, 5394-Temozolamide, 2578-Carmustine, 370-Gallic acid, 4497-Nimodipine, 2520-Verapamil, 5353639-Gambogic acid, 369312-Perrilyl alcohol. b Predicted apparent MDCK cell permeability in nm/sec (acceptable range: < 25 is poor, >500 is high ). c Predicted IC50 value for blockage of HERG K+ channels.(Concern below -5). d Predicted Caco-2 cell permeability in nm/s (acceptable range: < 25 is poor and > 500 is high). e Number of property or descriptor values that fall outside the 95% range of similar values for known drugs). f Number of violations of Lipinski's rule of five (maximum is 4).