In silico screening and molecular docking of bioactive agents towards human coronavirus receptor

Coronavirus infection has turned into pandemic despite of efforts of efforts of countries like America, Italy, China, France etc. Currently India is also outraged by the virulent effect of coronavirus. Although World Health Organisation initially claimed to have all controls over the virus, till date infection has coasted several lives worldwide. Currently we do not have enough time for carrying out traditional approaches of drug discovery. Computer aided drug designing approaches are the best solution. The present study is completely dedicated to in silico approaches like virtual screening, molecular docking and molecular property calculation. The library of 15 bioactive molecules was built and virtual screening was carried towards the crystalline structure of human coronavirus (6nzk) which was downloaded from protein database. Pyrx virtual screening tool was used and results revealed that F14 showed best binding affinity. The best screened molecule was further allowed to dock with the target using Autodock vina software. The results of docking of F14 with target using autodock vina revealed the binding affinity of -10.6. The interaction study with discovery studio visualize revealed the molecular interaction with serine, histidine, asparagine, leucine, tyrosine, lysine, isoleucine, threonine. The molecular properties of F14 were also calculated. The study helped us understand that 3-amino-2-phenylquinazolin4(3H)-one molecule can act as a potent parent structure for treating pandemic of coronavirus. This might be an essential tool for the medicinal chemist for designing novel molecule for treatment of coronavirus infection.


Introduction
Respiratory disorders are very common in country like India. The major reason for such disorders is pollution. The emissions from automobiles and several industries are degrading the quality of air. Currently the world is fighting against coronavirus disease [1].The origin of this virus is from china and has currently spread all over the world. COVID-19 is an infectious agent responsible for the disease. The coronavirus has entered the periphery of Indian borders through several migrants. The contagious and infectious nature of the virus has lead to spreading among several individuals [2].
People infected with the COVID-19 virus experiences mild to moderate respiratory illness, older people, and those with underlying medical problems like cardiovascular disease, diabetes, chronic respiratory disease, and cancer are more likely to develop serious illness.
At this time, there are no specific vaccines or treatments for COVID-19. However, there are many ongoing clinical trials evaluating potential treatments. WHO will continue to provide updated information as soon as clinical findings become available [3].
The exposure of corona virus has become very virulent in december 2019 and till date i.e. march 2020 and has claimed several lives worldwide. The number of infected persons isincreasing but still there are no vaccines or ideal medicines developed world wide for treating infection. Many countries have also lost control on this pandemic. The world health organization (WHO) has issued several guidelines which only includes details of prophylactic therapy. This prophylactic therapy includes sanitization as a key measure [4].
Computer aided drug designing is an innovative approach which utilizes in silico techniques to minimize the time and efforts in the pipeline of drug discovery. These techniques include study of the protein structure responsible for the disease or disorder, then identifying a ligand molecule that might have potent activities. Molecular docking techniques are study of interaction of the ligand with the protein molecule at active site. The active site is that three dimensional site of protein molecule which directly effects the activity of the protein. Although the activity can be either positive or negative but still it helps medicinal chemist to proceed on rational path [5][6].
The interaction ligands at receptor binding sites are measured as binding affinity. The less is binding affinity the more efficiently to the receptor at prescribed site. The charmm force fields are utilized for calculating these values. Although the values vary marginally differ from the system to system and software to software. This is due the difference in the genetic algorithm of the systems.

Avogadro software
Avogadro software was used to convert the .mol file to .pdb format [8]. This is again open source software that helps optimize the chemical structure. This also helps in minimizing the energy which is very important protocol for in silico studies. The software also allows generation of structure through SMILES or by drawing tool. These chemical structures were saved in the format of .pdb format which is required for docking purpose.

Pyrx software
Pyrx software was used for virtual screening of library of derivatives [9]. The pyrx software is a open softwarefor virtual screening. The approach includes blasting of several ligand molecules to a target and segregate the best fit molecules from the library. The tool includes a 4 step protocol and can screen a big library of molecules simultaneously on a defined site of target/receptor. The results can easily be exported in microsoft excel format as .csv file.

Autodock Vina software
Autodock vina software [10]was used to carry molecular docking and check the interactions. Autodock is a preliminary docking program and is used for the semi-flexible protein ligand docking studies. The software helps us to generate top 10 conformations of ligands which is arranged in decreasing order. The conformation at first position is said to be the best comformation with root mean square deviation (RMSD) value to be zero.
The crystalline structure of human coronavirus was downloaded from protein database and the pdb code was [PDB: 6nzk] [11][12]. Discovery studio 3.5 visualizer (DS visualizer) was used for molecular interaction and visualization [13].

Molinspiration online property calculator
Molinspiration online property calculator [14] was used to study the chemical properties of the compound. The best molecule which was screened from the library by Pyrx was docked using autodock software. The properties of the best molecule were calculated using online server of molinspiration software.

Preparation of ligand
Ligand structure was drawn using Chemsketch software and the structure was cleaned using the clean structure tool. The structure was saved in the working folder as .mol file in a working folder. The .mol file present in working folder was then accessed in Avogadro software and structure was optimized using optimization tool. The optimized structure was exported in the working folder as F1.pdb -F15.pdb file format.

Preparation of receptor
The crystal structure of human coronavirus was downloaded in .pdb format from the online database. The ligands which were previously attached at the active sites were removed using DS visualize. The treated target was then saved in the working folder as target.pdb. The results are tabulated in table 1

Receptor-Ligand Docking through Autodock vina
Autodock vina was used to identify binding poses with associated binding energies. As per the inverse relation of energy and stability, the conformation with greater binding energy is less stable. The Default parameters of the software program have been applied similar to the protocol followed elsewhere. The ligand and target molecule were converted into the .pdbqt format. Briefly, Lamarckian Genetic Algorithm (LGA) with default atomic salvation parameters 126 Å (x, y, and z) grid box in ratio of

Online chemical property calculator
Molinspiration online property calculator was used for calculating the properties of the ligand. The structure of the ligand was drawn using inbuilt tool and several properties were calculated. The properties were classified broadly into two types as structural property and bioactivity.

Virtual Screening
The results of virtual screening are tabulated below. The interactions with RMSD zero was considered to be best and the binding affinities of F1-F15 molecules are represented in table 1.

Molecular Docking
The best molecule selected from the virtual screening was F14. The result of docking of F14 with the target molecule is shown in Figure 3. Further visualization of interactions of ligand molecule was carried out using DS visualizer and the result is shown in Figure 4.

Figure 4 Interactions of ligand (F14) with target
The mean binding energy of the best conformation was found to be -10.6 and the root mean square deviation (RMSD) value was found to be zero.

Online chemical property calculator
The results of molinspiration online property calculator were represented as structural property and predicted bioactivity in table 2.

Conclusion
In silico screening of biologically active ligands were carried out towards human coronavirus. The library of 15 molecules were prepared and results of virtual screening revealed that F14 i.e. 3-amino-2-phenylquinazolin-4(3H)-one showed best binding affinity of -8.5. The results of docking of F14 with target using autodock vina revealed the binding affinity of -10.6 which was good. The interaction study revealed the molecular interaction with serine, histidine, asparagine, leucine, tyrosine, lysine, isoleucine and threonine. The molecular properties of F14 were also calculated.
The study helped us understand that 3-amino-2-phenylquinazolin-4(3H)-one molecule can act as a potent parent structure for treating pandemic of coronavirus. This might be an essential tool for the medicinal chemist [15] for designing novel molecule for treatment of coronavirus infection.
The lead molecule found to show best results can be further derivatized for development of potent bioactive agents.
Invivo studies can also be carried out for understanding the actual nature of molecule.