Pharmacophore modeling,multiple docking,and molecular dynamics studies on Wee1 kinase inhibitors

Wee1-like protein kinase (Wee1) is a tyrosine kinase that regulates the G₂ checkpoint and prevents entry into mitosis in response to DNA damage. Based on a series of signaling pathways initiated by Wee1, Wee1 has been recognized as a potential target for cancer therapy. To discover potent Wee1 inhibitors with novel scaffolds, ligand-based pharmacophore model has been built based on 101 known Wee1 inhibitors. Then the best pharmacophore model, AADRRR.340, with good partial least square (PLS) statistics (R²?=?0.9212, Q²?=?0.7457), was selected and validated. The validated model was used as a three-dimensional (3D) search query for databases virtual screening. The filtered molecules were further analyzed and refined by Lipinski’s rule of 5, multiple docking procedures (high throughput virtual screening (HTVS), standard precision (SP), genetic optimization for ligand docking (GOLD), extra precision (XP), and unique quantum polarized ligand docking (QPLD)); absorption, distribution, metabolism, excretion, and toxicity (ADMET) screening; and the Prime/molecular mechanics generalized born surface area (MM-GBSA) method binding free energy calculations. Eight leads were identified as potential Wee1 inhibitors, and a 50?ns molecular dynamics (MD) simulation was carried out for top four inhibitors to predict the stability of ligand–protein complex. Molecular mechanics Poisson–Boltzmann surface area (MM-PBSA) based on MD simulation and the energy contribution per residue to the binding energy were calculated. In the end, three hits with good stabilization and affinity to protein were identified.

Communicated by Ramaswamy H. Sarma     

Inhibitor design against JNK1 through e-pharmacophore modeling docking and molecular dynamics simulations

c-Jun-NH2 terminal kinases (JNKs) come under a class of serine/threonine protein kinases and are encoded by three genes, namely JNK1, JNK2 and JNK3. Human JNK1 is a cytosolic kinase belonging to mitogen-activated protein kinase (MAPK) family, which plays a major role in intracrinal signal transduction cascade mechanism. Overexpressed human JNK1, a key kinase interacts with other kinases involved in the etiology of many cancers, such as skin cancer, liver cancer, breast cancer, brain tumors, leukemia, multiple myeloma and lymphoma. Thus, to unveil a novel human JNK1 antagonist, receptor-based pharmacophore modeling was performed with the available eighteen cocrystal structures of JNK1 in the protein data bank. Eighteen e-pharmacophores were generated from the 18 cocrystal structures. Four common e-pharmacophores were developed from the 18 e-pharmacophores, which were used as three-dimensional (3D) query for shape-based similarity screening against more than one million small molecules to generate a JNK1 ligand library. Rigid receptor docking (RRD) performed using GLIDE v6.3 for the 1683 compounds from in-house library and 18 cocrystal ligands with human JNK1 from lower stringency to higher stringency revealed 17 leads. Further to derive the best leads, dock complexes obtained from RRD were studied further with quantum-polarized ligand docking (QPLD), induced fit docking (IFD) and molecular mechanics/generalized Born surface area (MM-GBSA). Four leads have showed lesser binding free energy and better binding affinity towards JNK1 compared to 18 cocrystal ligands. Additionally, JNK1–lead1 complex interaction stability was reasserted using 50?ns MD simulations run and also compared with the best resolute cocrystal structure using Desmond v3.8. Thus, the results obtained from RRD, QPLD, IFD and MD simulations indicated that lead1 might be used as a potent antagonist toward human JNK1 in cancer therapeutics.     

Multiple templates-based homology modeling enhances structure quality of AT1 receptor: validation by molecular dynamics and antagonist docking

We present a comparative account on 3D-structures of human type-1 receptor (AT1) for angiotensin II (AngII), modeled using three different methodologies. AngII activates a wide spectrum of signaling responses via the AT1 receptor that mediates physiological control of blood pressure and diverse pathological actions in cardiovascular, renal, and other cell types. Availability of 3D-model of AT1 receptor would significantly enhance the development of new drugs for cardiovascular diseases. However, templates of AT1 receptor with low sequence similarity increase the complexity in straightforward homology modeling, and hence there is a need to evaluate different modeling methodologies in order to use the models for sensitive applications such as rational drug design. Three models were generated for AT1 receptor by, (1) homology modeling with bovine rhodopsin as template, (2) homology modeling with multiple templates and (3) threading using I-TASSER web server. Molecular dynamics (MD) simulation (15 ns) of models in explicit membrane-water system, Ramachandran plot analysis and molecular docking with antagonists led to the conclusion that multiple template-based homology modeling outweighs other methodologies for AT1 modeling.     

Computational investigation of the substrate recognition mechanism of protein D-aspartyl (L-isoaspartyl) O-methyltransferase by docking and molecular dynamics simulation studies and application to interpret size exclusion chromatography data

Unusual amino acid residues such as L-β-aspartyl (Asp), D-α-Asp, and D-β-Asp have been detected in proteins and peptides such as α-crystallin in the lens and β-amyloid in the brain. These residues increase with age, and hence they are associated with age-related diseases. The enzyme protein D-aspartyl (L-isoaspartyl) O-methyltransferase (PIMT) can revert these residues back to the normal L-α-Asp residue. PIMT catalyzes transmethylation of S-adenosylmethionine to L-β-Asp and D-α-Asp residues in proteins and peptides. In this work, the substrate recognition mechanism of PIMT was investigated using docking and molecular dynamics simulation studies. It was shown that the hydrogen bonds of Ser60 and Val214 to the carboxyl group of Asp are important components during substrate recognition by PIMT. In addition, specific hydrogen bonds were observed between the main chains of the substrates and those of Ala61 and Ile212 of PIMT when PIMT recognized L-β-Asp. Hydrophobic interactions between the (n-1) residue of the substrates and Ile212 and Val214 of PIMT may also have an important effect on substrate binding. Volume changes upon substrate binding were also evaluated in the context of possible application to interpretation of size exclusion chromatography data.     

Molecular modeling studies of atorvastatin analogues as HMGR inhibitors using 3D-QSAR,molecular docking and molecular dynamics simulations

3-Hydroxy-3-methylglutaryl coenzyme-A reductase (HMGR) is generally regarded as targets for the treatment of hypercholesterolemia. HMGR inhibitors (more commonly known as statins) are discovered as plasma cholesterol lowering molecules. In this work, 120 atorvastatin analogues were studied using a combination of molecular modeling techniques including three-dimensional quantitative structure–activity relationship (3D-QSAR), molecular docking and molecular dynamics (MD) simulation. The results show that the best CoMFA (comparative molecular field analysis) model has q² = 0.558 and r² = 0.977, and the best CoMSIA (comparative molecular similarity indices analysis) model has q² = 0.582 and r² = 0.919. Molecular docking and MD simulation explored the binding relationship of the ligand and the receptor protein. The calculation results indicated that the hydrophobic and electrostatic fields play key roles in QSAR model. After MD simulation, we found four vital residues (Lys735, Arg590, Asp690 and Asn686) and three hydrophobic regions in HMGR binding site. The calculation results show that atorvastatin analogues obtained by introduction of F atoms or gem-difluoro groups could obviously improve the inhibitory activity. The new HMGR inhibitor analogues design in this Letter had been submitted which is being currently synthesized by our laboratories.     

Homology modeling,molecular dynamic simulation and docking studies of cyclin dependent kinase 1

In order to develop promising cyclin dependent kinase 1 inhibitors, homology modeling, docking and molecular dynamic simulation techniques were applied to get insight into the functional and structural properties of cyclin dependent kinase 1 (CDK1). Since there is no reported CDK1 crystal structural data, the three dimensional structure of CDK1 was constructed based on homology modeling. An extensive dynamic simulation was also performed on a Flavopiridol-CDK1 complex for probing the binding pattern of Flavopiridol in the active site of CDK1. The binding modes of other inhibitors to CDK1 were also proposed by molecular docking. The structural requirement for developing more potent CDK1 inhibitors was obtained by the above-mentioned molecular simulations and pharmacophore modeling.     

Deciphering the GPER/GPR30-agonist and antagonists interactions using molecular modeling studies,molecular dynamics,and docking simulations

The G-protein coupled estrogen receptor 1 GPER/GPR30 is a transmembrane seven-helix (7TM) receptor involved in the growth and proliferation of breast cancer. Due to the absence of a crystal structure of GPER/GPR30, in this work, molecular modeling studies have been carried out to build a three-dimensional structure, which was subsequently refined by molecular dynamics (MD) simulations (up to 120 ns). Furthermore, we explored GPER/GPR30’s molecular recognition properties by using reported agonist ligands (G1, estradiol (E2), tamoxifen, and fulvestrant) and the antagonist ligands (G15 and G36) in subsequent docking studies. Our results identified the E2 binding site on GPER/GPR30, as well as other receptor cavities for accepting large volume ligands, through GPER/GPR30 π–π, hydrophobic, and hydrogen bond interactions. Snapshots of the MD trajectory at 14 and 70 ns showed almost identical binding motifs for G1 and G15. It was also observed that C107 interacts with the acetyl oxygen of G1 (at 14 ns) and that at 70 ns the residue E275 interacts with the acetyl group and with the oxygen from the other agonist whereas the isopropyl group of G36 is oriented toward Met141, suggesting that both C107 and E275 could be involved in the protein activation. This contribution suggest that GPER1 has great structural changes which explain its great capacity to accept diverse ligands, and also, the same ligand could be recognized in different binding pose according to GPER structural conformations.     

Computational screening of dipeptidyl peptidase IV inhibitors from micoroalgal metabolites by pharmacophore modeling and molecular docking

Dipeptidyl peptidase IV (DPP‐IV) catalyzes conversion of GLP‐1 (glucagon like peptide 1) to inert structure, which results in insufficient secretion of insulin and increase in postprandial blood glucose level. The present study attempts to identify novel inhibitors from bioactive metabolites present in microalgae against DPP‐IV through virtual screening, molecular docking, and pharmacophore modeling for the active target. Possible binding modes of all 60 ligands against DPP‐IV receptor were constructed using MTiOpenScreen virtual screening server. Pharmacophore model was built based on identified 38 DPP‐IV test ligands by using the web‐based PharmaGist program which encompasses hydrogen‐bond acceptors, hydrophobic groups, spatial features, and aromatic rings. The pharmacophore model having highest scores was selected to screen active DPP‐IV ligands. Highest scoring model was used as a query in ZincPharmer screening. All identified ligands were filtered, based on the Lipinski's rule‐of‐five and were subjected to docking studies. In the process of docking analyses, we considered different bonding modes of one ligand with multiple active cavities of DPP‐IV with the help of AutoDock 4.0. The docking analyses indicate that the bioactive constituents, namely, β‐stigmasterol, barbamide, docosahexaenoic acid, arachidonic acid, and harman showed the best binding energies on DPP‐IV receptor and hydrogen bonding with ASP545, GLY741, TYR754, TYR666, ARG125, TYR547, SER630, and LYS554 residues. This study concludes that docosahexaenoic acid, arachidonic acid, β‐stigmasterol, barbamide, harman, ZINC58564986, ZINC56907325, ZINC69432950, ZINC69431828, ZINC73533041, ZINC84287073, ZINC69849395, and ZINC10508406 act as possible DPP‐IV inhibitors.     

Insights into the structure and inhibition of Giardia intestinalis arginine deiminase: homology modeling,docking, and molecular dynamics studies

Giardia intestinalis arginine deiminase (GiADI) is an important metabolic enzyme involved in the energy production and defense of this protozoan parasite. The lack of this enzyme in the human host makes GiADI an attractive target for drug design against G. intestinalis. One approach in the design of inhibitors of GiADI could be computer-assisted studies of its crystal structure, such as docking; however, the required crystallographic structure of the enzyme still remains unresolved. Because of its relevance, in this work, we present a three-dimensional structure of GiADI obtained from its amino acid sequence using the homology modeling approximation. Furthermore, we present an approximation of the most stable dimeric structure of GiADI identified through molecular dynamics simulation studies. An in silico analysis of druggability using the structure of GiADI was carried out in order to know if it is a good target for design and optimization of selective inhibitors. Potential GiADI inhibitors were identified by docking of a set of 3196 commercial and 19 in-house benzimidazole derivatives, and molecular dynamics simulation studies were used to evaluate the stability of the ligand–enzyme complexes.     

A combination of pharmacophore modeling,atom-based 3D-QSAR,molecular docking and molecular dynamics simulation studies on PDE4 enzyme inhibitors

Phosphodiesterases 4 enzyme is an attractive target for the design of anti-inflammatory and bronchodilator agents. In the present study, pharmacophore and atom-based 3D-QSAR studies were carried out for pyrazolopyridine and quinoline derivatives using Schrödinger suite 2014-3. A four-point pharmacophore model was developed using 74 molecules having pIC₅₀ ranging from 10.1 to 4.5. The best four feature model consists of one hydrogen bond acceptor, two aromatic rings, and one hydrophobic group. The pharmacophore hypothesis yielded a statistically significant 3D-QSAR model, with a high correlation coefficient (R^2?=?.9949), cross validation coefficient (Q^2?=?.7291), and Pearson-r (.9107) at six component partial least square factor. The external validation indicated that our QSAR model possessed high predictive power with R² value of .88. The generated model was further validated by enrichment studies using the decoy test. Molecular docking, free energy calculation, and molecular dynamics (MD) simulation studies have been performed to explore the putative binding modes of these ligands. A 10-ns MD simulation confirmed the docking results of both stability of the 1XMU–ligand complex and the presumed active conformation. Outcomes of the present study provide insight in designing novel molecules with better PDE4 inhibitory activity.     

Computational modeling of human coreceptor CCR5 antagonist as a HIV-1 entry inhibitor: using an integrated homology modeling,docking, and membrane molecular dynamics simulation analysis approach

Chemokine receptor 5 (CCR5) is an integral membrane protein that is utilized during human immunodeficiency virus type-1 entry into host cells. CCR5 is a G-protein coupled receptor that contains seven transmembrane (TM) helices. However, the crystal structure of CCR5 has not been reported. A homology model of CCR5 was developed based on the recently reported CXCR4 structure as template. Automated docking of the most potent (14), medium potent (37), and least potent (25) CCR5 antagonists was performed using the CCR5 model. To characterize the mechanism responsible for the interactions between ligands (14, 25, and 37) and CCR5, membrane molecular dynamic (MD) simulations were performed. The position and orientation of ligands (14, 25, and 37) were found to be changed after MD simulations, which demonstrated the ability of this technique to identify binding modes. Furthermore, at the end of simulation, it was found that residues identified by docking were changed and some new residues were introduced in the proximity of ligands. Our results are in line with the majority of previous mutational reports. These results show that hydrophobicity is the determining factor of CCR5 antagonism. In addition, salt bridging and hydrogen bond contacts between ligands (14, 25, and 37) and CCR5 are also crucial for inhibitory activity. The residues newly identified by MD simulation are Ser160, Phe166, Ser180, His181, and Trp190, and so far no site-directed mutagenesis studies have been reported. To determine the contributions made by these residues, additional mutational studies are suggested. We propose a general binding mode for these derivatives based on the MD simulation results of higher (14), medium (37), and lower (25) potent inhibitors. Interestingly, we found some trend for these inhibitors such as, salt bridge interaction between basic nitrogen of ligand and acidic Glu283 seemed necessary for inhibitory activity. Also, two aromatic pockets (pocket I – TM1-3 and pocket II – TM3-6) were linked by the central polar region in TM7, and the simulated inhibitors show important interactions with the Trp86, Tyr89, Tyr108, Phe112, Ile198, Tyr251, Leu255, and Gln280 and Glu283 residues. These results shed light on the usage of MD simulation to identify more stable, optimal binding modes of the inhibitors.     

Molecular docking,molecular modeling,and molecular dynamics studies of azaisoflavone as dual COX-2 inhibitors and TP receptor antagonists

Designed multi-target ligand (DML) is an emerging strategy for the development of new drugs and involves the engagement of multiple targets with the same moiety. In the context of NSAIDs it has been suggested that targeting the thromboxane prostanoid (TP) receptor along with cyclooxygenase-2 (COX-2) may help to overcome cardiovascular (CVS) complications associated with COXIBs. In the present work, azaisoflavones were studied for their COX-2 and TP receptor binding activities using structure based drug design (SBDD) techniques. Flavonoids were selected as a starting point based on their known COX-2 inhibitory and TP receptor antagonist activity. Iterative design and docking studies resulted in the evolution of a new class scaffold replacing the benzopyran-4-one ring of flavonoids with quinolin-4-one. The docking and binding parameters of these new compounds are found to be promising in comparison to those of selective COX-2 inhibitors, such as SC-558 and celecoxib. Owing to the lack of structural information, a model for the TP receptor was generated using a threading base alignment method with loop optimization performed using an ab initio method. The model generated was validated against known antagonists for TP receptor using docking/MMGBSA. Finally, the molecules that were designed for selective COX-2 inhibition were docked into the active site of the TP receptor. Iterative structural modifications and docking on these molecules generated a series which displays optimum docking scores and binding interaction for both targets. Molecular dynamics studies on a known TP receptor antagonist and a designed molecule show that both molecules remain in contact with protein throughout the simulation and interact in similar binding modes.

Open image in new window

Graphical abstract ?

     

Pharmacophore modeling,molecular docking and molecular dynamics studies on natural products database to discover novel skeleton as non-purine xanthine oxidase inhibitors

Gout is a common inflammatory arthritis caused by the deposition of urate crystals within joints. It is increasingly in prevalence during the past few decades as shown by the epidemiological survey results. Xanthine oxidase (XO) is a key enzyme to transfer hypoxanthine and xanthine to uric acid, whose overproduction leads to gout. Therefore, inhibiting the activity of xanthine oxidase is an important way to reduce the production of urate. In the study, in order to identify the potential natural products targeting XO, pharmacophore modeling was employed to filter databases. Here, two methods, pharmacophore based on ligand and pharmacophore based on receptor-ligand, were constructed by Discovery Studio. Then GOLD was used to refine the potential compounds with higher fitness scores. Finally, molecular docking and dynamics simulations were employed to analyze the interactions between compounds and protein. The best hypothesis was set as a 3D query to screen database, returning 785 and 297 compounds respectively. A merged set of the above 1082 molecules was subjected to molecular docking, which returned 144 hits with high-fitness scores. These molecules were clustered in four main kinds depending on different backbones. What is more, molecular docking showed that the representative compounds established key interactions with the amino acid residues in the protein, and the RMSD and RMSF of molecular dynamics results showed that these compounds can stabilize the protein. The information represented in the study confirmed previous reports. And it may assist to discover and design new backbones as potential XO inhibitors based on natural products.     

Constant pH molecular dynamics (CpHMD) and molecular docking studies of CquiOBP1 pH-induced ligand releasing mechanism

The odorant binding protein of Culex quinquefasciatus (CquiOBP1), expressed on the insect antenna, is crucial for the investigation of trapping baited with oviposition semi-chemicals and controlling mosquito populations. The acidic titratable residues pKa prediction and the ligand binding poses investigation in two systems (pH 7 and pH 5) are studied by constant pH molecular dynamics (CpHMD) and molecular docking methods. Research results reveal that the change of the protonation states would disrupt some important H-bonds, such as Asp 66-Asp 70, Glu 105-Asn 102, etc. The cleavage of these H-bonds leads to the movement of the relative position of hydrophobic tunnel, N- and C- termini loops and pH-sensing triad (His23-Tyr54-Val125) in acid solution. Ligand MOP has lower affinity and shows different binding poses to protein CquiOBP1 at pH 5. This ligand may be released from another tunnel between helices α3 and α4 in acidic environment. However, it would bind to the protein with high affinity in neutral environment. This work could provide more penetrating understanding of the pH-induced ligand-releasing mechanism.     

Computational study of EGFR inhibition: molecular dynamics studies on the active and inactive protein conformations

The structural diversity observed across protein kinases, resulting in subtly different active site cavities, is highly desirable in the pursuit of selective inhibitors, yet it can also be a hindrance from a structure-based design perspective. An important challenge in structure-based design is to better understand the dynamic nature of protein kinases and the underlying reasons for specific conformational preferences in the presence of different inhibitors. To investigate this issue, we performed molecular dynamics simulation on both the active and inactive wild type epidermal growth factor receptor (EGFR) protein with both type-I and type-II inhibitors. Our goal is to better understand the origin of the two distinct EGFR protein conformations, their dynamic differences, and their relative preference for Type-I inhibitors such as gefitinib and Type-II inhibitors such as lapatinib. We discuss the implications of protein dynamics from a structure-based design perspective.     

Binding efficiencies of carbohydrate ligands with different genotypes of cholera toxin B: molecular modeling, dynamics and docking simulation studies

Vibrio cholerae produces cholera toxin (CT) that consists of two subunits, A and B, and is encoded by a filamentous phage CTXΦ. The A subunit carries enzymatic activity that ribosylates ADP, whereas the B subunit binds to monosialoganglioside (GM1) receptor in epithelial cells. Molecular analysis of toxigenic V. cholerae strains indicated the presence of multiple ctxB genotypes. In this study, we employed a comparative modeling approach to define the structural features of all known variants of ctxB found in O139 serogroup V. cholerae. Modeling, molecular dynamics and docking simulations studies suggested subtle variations in the binding ability of ctxB variants to carbohydrate ligands of GM1 (galactose, sialic acid and N-acetyl galactosamine). These findings throw light on the molecular efficiencies of pathogenic isolates of V. cholerae harboring natural variants of ctxB in causing the disease, thus suggesting the need to consider ctxB variations when designing vaccines against cholera.     

3D QSAR CoMFA/CoMSIA, molecular docking and molecular dynamics studies of fullerene-based HIV-1 PR inhibitors

For the first time, a set of experimentally reported [60] fullerene derivatives were subjected to the 3D-QSAR/CoMFA and CoMSIA studies. The aim of this study is to propose a series of novel [60] fullerene-based inhibitors with optimal binding affinity for the HIV-1 PR enzyme. The position of the template molecule at the cavity of HIV-1 PR was optimized and 3D QSAR models were developed. Relative contributions of steric/electrostatic fields of the 3D-QSAR/CoMFA and CoMSIA models have shown that steric effects govern the bioactivity of the compounds, but electrostatic interactions play also an important role.The de novo drug design Leapfrog simulations provided a series of novel compounds with predicted improved inhibition effect.     

Synergistic effects in the designs of neuraminidase ligands: Analysis from docking and molecular dynamics studies

Docking and molecular dynamics were used to study the nine ligands (see Scheme 1) at the neuraminidase (NA) active sites. Their binding modes are structurally and energetically different, with details given in the text. Compared with 1A (oseltamivir carboxylate), the changes of core template or/and functional groups in the other ligands cause the reductions of interaction energies and numbers of H-bonds with the NA proteins. Nonetheless, all these ligands occupy the proximity space at the NA active sites and share some commonness in their binding modes. The fragment approach was then used to analyze and understand the binding specificities of the nine ligands. The contributions of each core template and functional group were evaluated. It was found that the core templates rather than functional groups play a larger role during the binding processes; in addition, the binding qualities are determined by the synergistic effects of the core templates and functional groups. Among the nine ligands, 1A (oseltamivir carboxylate) has the largest synergistic energy and its functional groups fit perfectly with the NA active site, consistent with the largest interaction energy, numerous H-bonds with the NA active-site residues as well as experimentally lowest IC₅₀ value. Owing to the poorer metabolizability than oseltamivir, large contribution of the benzene core template and fine synergistic effects of the functional groups, the 4-(N-acetylamino)-5-guanidino-3-(3-pentyloxy)benzoic acid should be an ideal lead compound for optimizing NA drugs.     

Identification of novel urease inhibitors: pharmacophore modeling,virtual screening and molecular docking studies

Abstract

Pharmacophore modeling and atom-based three-dimensional quantitative structure–activity relationship (3D-QSAR) have been developed on N-acylglycino- and hippurohydroxamic acid derivatives, which are known potential inhibitors of urease. This is followed by virtual screening and ADMET (absorption, distribution, metabolism, excretion and toxicity) studies on a large library of known drugs in order to get lead molecules as Helicobacter pylori urease inhibitors. A suitable three-featured pharmacophore model comprising one H-bond acceptor and two H-bond donor features (ADD.10) has been found to be the best QSAR model. An external library of compounds (～3000 molecules), pre-filtered using Lipinski’s rule of five, has been further screened using the pharmacophore model ADD.10. By analyzing the fitness of the hits with respect to the pharmacophore model and their binding interaction inside the urease active site, four molecules have been predicted to be extremely good urease inhibitors. Two of these have significant potential and should be taken up for further drug-designing process.