br Conflicts of interest br Financial support This work

Conflicts of interest
Financial support This work was supported by grants from the Fondazione Cariplo [Grant number 2011-0463] (Carini); and by Funds for Original Research of the Università del Piemonte Orientale (2016, Project: Carini-Boldorini). The sponsors had no involvement in study design; in the collection, analysis and interpretation of data; in the writing of the report; and in the decision to submit the article for publication.
Introduction
Results
Discussion
Experimental procedure
Conflicts of interest
Acknowledgments This research was supported by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Education, Science, and Technology (NRF-2014R1A2A2A01006556).
Introduction The feeling, “Drug development process is time- and cost-intensive”, poses a big challenge itself, hold at bay virtually all publicly funded research institutions and most of the private sector pharma companies from venturing into it. It carries on average 25 years with the estimated life-cycle R&D cost of $1.7 to $2 billion per approved glibenclamide (Chakravarthy et al., 2015). To cut it short, nowadays, in silico based approaches are being adopted into consideration from the beginning (Lavecchia and Giovanni, 2013; Tanrikulu et al., 2013; Lounnas et al., 2013). Foremost of all, suitable drug targets are identified, such as enzymes, transporters, receptors. These marks may be endogenous, which requires modulation for desired therapeutic outcome or of exogenous origin, which is present in pathogen and is essential for its endurance. In the following stride, three-dimensional (3D) structure is taken, preferably determined through X-ray crystallography, NMR, or modeled using homology modeling approach in the case of unavailability of the crystal structure. Subsequently, Structure Based Virtual Screening (SBVS) is applied and screened hits are subjected to experimental proof. There is also a consistent decline observed in a number of small molecule NCEs in the period of 1981–2010 (Newman and Cragg, 2012). Likewise, to exploit the loose nature of the compound and increase its therapeutic efficacy a new prototype “one drug multiple targets”, also known as polypharmacology is gaining importance recently (Reddy and Zhang, 2013).
Materials and method Protein sequence was retrieved from the UniProt database (www.uniprot.org) (Q5AMH1). I-Tasser v4.4 suite was used for protein structure modeling with 15 iterative threading assembly simulations extending 50 h each (Yang et al., 2015). The authenticity of the developed protein model was checked by PROCHECK (Laskowski et al., 1993). Chemical samples were retrieved from PubChem compounds database (NCBI, USA) (https://pubchem.ncbi.nlm.nih.gov/). Their PubChem compounds fingerprints were extracted using PaDEL-descriptor tool (Yap, 2011). Subsequently, these fingerprints were clustered based on their Tanimoto similarity index (Haque et al., 2011). A representative chemical sample from each cluster was drawn randomly to build virtual screening (VS I) library. Then, Vina was used to screen newly built library against complete modeled structure (Trott and Olson, 2010). Later, Active Domain (AD) of the modeled protein was selected for further studies on the ground of a number of bound ligands and their binding affinities towards particular domain. All the molecular dynamics (MD) simulations were performed using GROMACS 5.1 (Abraham et al., 2015) with Amber99sb-ildn force field (Lindorff-Larsen et al., 2010); dodecahedron box with minimal distance of 1 nm between protein surface and edge of the box; SPC/E water model (Berendsen et al., 1987); inclusion of counter ions; energy minimization using steepest descent; LINCS algorithm for all bonds constrain (Hess et al., 1997); Verlet cutoff-scheme using grid method with short-range nonbonded interaction cutoff of 1 nm (Páll and Hess, 2013); Particle Mesh Ewald (PME) for long-range electrostatics with fourth-order cubic interpolation and 1.6 Å grid spacing (Essmann et al., 1995); and, velocity-rescale thermostat (Bussi et al., 2007) and Parrinello-Rahman barostat (Parrinello and Rahman, 1981) coupling methods to maintain temperature and pressure of the system at 300 K and 1 bar, respectively. The system was equilibrated for 1 ns NVT followed by 1 ns NPT run with position restraints on protein and ligand (in the case of protein-ligand complex). Production MD simulation on AD was carried out for 300 ns, and conformation saved at every 2 ps. To get native conformation of AD, free energy landscape (FEL) analysis was done on the ensemble structure obtained during the course of the simulation. The true binding site on the surface of native AD was delineated through the second round of VS (II) experiment. Later, again random extraction of a representative ligand from each cluster was done in the building of new VS library (II).