SARS-COV-2 Spike Glycoprotein as Inhibitory Target for Insilico Screening of Natural Compounds

Coronavirius disease 2019 (Covid-19) pandemic caused by SARS-Cov-2 has raised global health concern without approved drug for management of this lie threatening disease. The aim of this study was to predict the inhibitory potential of quercetin-3-o-rutinoside against SARS-Cov-2 spike glycoprotein. Targeting the SARS-Cov-2 spike protein from angiotensin converting enzyme 2 complex (pdb: 6lzg) is gaining importance. In this study, in silico computational relationship between plant-derived natural drug and spike glycoprotein was predicted. The results were evaluated based on glide (Schrodinger) dock score. Among the five (5) screened compounds, quercetin-3-o-rutinoside has the best docking score (-9.296) with the target. Molecular dynamic (MD) simulation study was performed for 1000ps to confirm the stability behavior of the spike protein and quercetin-3-o-rutinoside complex. The MD simulation study validated the stability of quercetin-3-o-rutinoside in the spike protein binding pocket as potent inhibitor.


Introduction
Coronavirus (Cov) belongs to the family coronaviridae, considered to be the largest RNA virus with genomes ranging from 27 to 32kb (Masters, 2016;Rahman et al., 2020). Coronavirus disease 2019 (Covid-19) is a major life threatening disease worldwide due to its fast spreading (Kadioglul et al., 2020). It is caused by severe acute respiratory syndrome coronavirus 2 (SARS-Cov-2) (Omotuyi et al., 2020). The SARS-Cov-2 causes mainly severe acute and other complications in the heart, kidney, brain and spleen ultimately leading to disastrous effect in Covid-19 afflicted patients (Madjid et al., 2020). SARS-Cov-2 invade human cells through its spike proteins interaction with angiotensin converting enzyme 2 (ACE 2) receptors. As a structural glycoprotein on the virion surface, the spike glycoprotein of coronavirus binds to the host cellular receptors followed by fusion between the viral envelope and the cellular membrane (Zhang et al., 2005). Upon binding with the receptor, the protein changes its conformation from a pre-fusion to a post-fusion form (Pandey et al., 2020).
The viral glycosylated spike protein composed of S1 and S2 subunits in the coronavirus which enables them access into the host cell. The receptor bonding domain of S1 subunit in SARS-CoV-2 attaches to ACE2, led to the shed of S1 subunit and subsequently triggers the cleavage of the S2 subunit by the host protease protein (Hoffman et al., 2020). The biophysical and structural evidence shows that, SARS-CoV-2 spike protein binds with ACE2 receptors with higher affinity (Patel et al., 2020). Therefore, it is a target for small molecules to generate potential inhibitors of ACE2 interaction with the protein.
In silico studies of small molecules including those of natural compounds of plant origin have been screened and confirm to directly inhibit this important protein of SARS-Cov-2 (Zhang et al., 2020;Chen et al., 2020;Wen et al., 2007). Although drug development against coronavirus (Covid-19) has poses great challenge to scientist. However, the virtual screening of databases and the use of bioinformatics and cheminformatics have improved the discovery of small molecules that binds to the selected target which is the preliminary phase of drug design. Here, five natural compounds were screened, targeting the entry pathway of SARS-Cov-2 penetration into cells. The Quercetin and Kaempferol derivatives screened in this study are widely distributed plant flavonoids, found in several vegetables, seeds, grains (Biancatelli et al., 2020) and other plant parts. Studies suggested that these compounds promote antioxidant Quercetin has been studied in various models of viral infection due to its antiviral efficacy through inhibition of protease, ploymerases, DNA gyrase suppression and reverse transcriptase inhibition (Biancatelli et al., 2020).
In this in silico study, molecular docking and molecular dynamics simulation were performed to ascertain the most potent inhibitory compounds against SARS-CoV-2 spike protein. The pharmacokinetics properties of these natural compounds were also explored to ascertain these compounds as therapeutic agents.

Materials and method
Glide tool from Schrodinger molecular drug discovery suite (version 2017-1) and AMBER 16 package were used for molecular docking and MD simulation respectively in this research.

Ligand Preparation
Five (5) phyto-compounds of rutin derivatives used in this research were obtained from published literatures and they were used to prepare the library for this research. The structures of the phytochemicals were downloaded in the Structure data format (sdf format) from the PubChem database (https://pubchem.ncbi.nlm.nih.gov). The ligands were prepared by LigPrep module of Maestro 11.5 interfaces in the Schrodinger suite 2017-1. They were converted from 2D to 3D structures by including stereo chemical, ionization, tautomeric variations, as well as energy minimization and optimized for their geometry, desalted and corrected for their chiralities and missing hydrogen atoms. The bonds orders of these ligands were fixed and the charged groups were neutralized. The ionization and tautomeric states were generated between pH of 7.0 +_ 2.0 using Epik module. In the LigPrep module, the compounds were minimized by Optimized Potentials for Liquid Simulations (OPLS3) force field. A single low energy ring confirmation per ligand was generated and the optimized ligands were used for docking analysis (Balogun et al., 2020).

Protein Preparation
Nucleocapsid spike glycoprotein crystal structure of SARS-CoV-2 was retrieved from Protein Data Bank (http://www.rcsb.org/pdb/home.do) (pdb ID: 6M3M). The maestro11.5 interface was used to view the 3D structure of SARS-CoV-2 nucleocapsid spike glycoprotein. The protein was prepared according to Balogun et al. (2020).

Nucleocapsid spike glycoprotein grid Generation
Using the module for generating receptor grid on maestro 11.5, the area of interaction between the nucleocapsid spike glycoprotein and ligands were generated. The centroid of the docked pose, in the binding box dimension of the spike protein, was patterned in terms of the coordinates x, y and z.

Molecular docking using glide tool
Using the prepared spike glycoprotein and ligands, the molecular docking of the molecules was

Pharmacokinetics screening of hit Compounds
The ADME toxicity depicts the absorption, distribution, metabolism, excretion, toxicity, and elucidates the pharmacokinetics profiles of the natural compounds. The ADME toxicity screening of the natural compounds were performed with the admetSAR prediction online tool

Molecular Dynamics Simulations
The molecular dynamics simulation was conducted at a simulation time of 10000 ps using the AMBER 16 package. The topology files of the superimposed complex were constructed using the in house program of the package. The antechamber program was used to assemble the force fields for the receptor residues and ligand molecules, while for the construction of the biopolymers from the component residues and preparation of the force field, the LEaP plugin    Therefore, co-administration of other drugs does not posse drug-drug interaction or toxicity.
From table 2, all the natural compounds are non-carcinogenic and non-toxic compounds.

Molecular Dynamics of protein-ligand complex
The    (4) hydrogen atom were obsered interacting with quercetin-3-o-rutinoside (Figure 5a). The RSMD crystallography resolution of the docked complex, ranged from 1.5 to 2.9Å at a simulation time of 1000ps, which signifies that both relationship are linear (Fig 5b). A resolution outside this range, denotes that the complex can't be linear. The protein had a conformational change as

Energy Calculations between the Ligand and the Protein receptor
The spike protein residues that contributes to the decompose results comprising of the electrostatic energy as calculated by the MM force field (TELE), van der Waals contribution from MM (TVDW), total gas phase energy (TGAS), sum of non-polar and polar contributions to solvation (TGBSOL), final estimated binding free energy calculated from the terms above (TGBTOT), are displayed in a heat map.The unit of them is kcal/mol (Fig. 8).

Principal Cluster Analysis (PCA)
The PCA tool quantitatively evaluates the collective motion and measures the moment direction (Sarma et al., 2020;Pandey et al., 2020). The PCA and dynamic cross-correlation analysis were performed to evaluate the allosteric effect of SARS-CoV-2 spike protein under the ligand binding based on MD trajectory. This is to assess the direct and efficient mechanism for the modulation and regulation of cellular function in response to changes in concentration of small molecules. Therefore, 50% of the total variance of a given protein family structure can be captured by the principal components. This clearly provides a considerable insight on the nature of conformational differences. From the two clusters, the differential conformations can be studied. The value of both clusters is 27.1%. Hence, the PCA result of the spike protein and the ligand had no significant allostery ( Fig. 9a and 9b).
(9a) (9b) Fig. 9a and 9b: The PCA cluster of conformations and residues cross correlation of the spike protein and ligand. The trajectory frames are highlighted in blue to white to red in time order. The N terminus is represented in blue, while the C-terminus is highlighted in red. The pink represents a pair of residues with correlation coefficient >0.8, and the blue lines depict a pair of residues with correlation coefficient <−0.4.

CONCLUSION
This work has identified quercetin-3-o-rutinoside as a direct inhibitor of SARS-CoV-2 spike protein through molecular docking analysis and prevents the interaction between human ACE-2 and SARS-CoV-2 spike protein which formed the key event of viral penetration and infection.
The pharmacodynamics profile of quercetin-3-o-rutinoside depicted from the ADME toxicity reveal that, this compound is a safe therapeutic agent while the MD simulation predicted the stable bound or interaction of quercetin-3-o-rutinoside with the binding pocket of SARS-CoV-2 spike protein.

Ethical statement
The study does not involve the animals.

Declaration of competing interest
The authors declare no competing interests.