In silico screening of natural compounds to inhibit interaction of human ACE2 receptor and spike protein of SARS-CoV-2 for the prevention of COVID-19

Abstract A computational investigation was carried out to find out potential phytochemicals that could inhibit the binding of human angiotensin-converting enzyme-2 (ACE2) receptors to spike protein of SARS-CoV-2 which is an essential step to gain entry inside human cells and onset of viral infection known as Coronavirus disease (COVID-19). A library of phytochemicals was screened by virtual screening against ACE2 receptors resulting in twenty phytochemicals out of 686 which had binding energy (−11.8 to −6.9 kcal/mol). Drug-likeness gave five hits, but ADMET analysis yielded 4 nontoxic hit phytochemicals. Molecular dynamics simulation of four-hit compounds resulted in acceptable stability and good dynamics behavior. These phytochemicals are Hinokinin, Gmelanone, Isocolumbin, and Tinocordioside, from Vitis vinifera, Gmelina arborea, and Tinospora cordifolia. The above-mentioned phytochemicals may be promising ACE2 inhibitors and can prevent infection of SARS-CoV-2 by inhibiting the entry of the virus into host cells. Communicated by Ramaswamy H. Sarma

Currently, many vaccines of coronaviruses are available in the world but no appropriate medicines are available to cure coronavirus infections, therefore it is subject of interest to discover potential drug candidates against coronavirus. In this regard, restricting the entry of the virus into the human body and boosting up the immune system may be synergetic ways to prevent the disease. In India, many plants and herbal formulations are documented to increase immunity, and their compounds can be screened against the coronavirus. Chyawanprash, a famous herbal formulation is prepared by processing around 50 such important medicinal herbs which possess multiple health benefits, including boosting the immune system for all age groups (Parle & Bansal, 2006).
The entry of SARS-CoV-2 can be inhibited by targeting the human ACE2 receptor. It is a functional receptor for spike protein of SARS-CoV-2 which facilitates the entry of virus in host cells. It has been reported that human-to-human transmission of SARS-CoV occurs by the binding between the receptor-binding domain (RBD) of virus spikes protein and ACE2 receptor (Jaimes et al., 2020). The spike protein of SARS-CoV-2 binds to ACE2 receptors in the lower respiratory tract of infected patients to gain entry into the lungs (Wan et al., 2020). ACE2 modulates the many activities of a protein called angiotensin II (ANG II) that enhance damage to body tissues and various types of tissue injuries. ACE2 is found in multiple cell types and tissues i.e., lungs, blood vessels in the heart, kidneys, liver, and gastrointestinal tract. It is located in epithelial cells that line some membranes and build defensive barriers. The viral infection can be stopped by blocking the virus entry into the host cell (Purohit et al., 2011;Purohit & Sethumadhavan, 2009) (Figure 1). The blockage of the ACE2 catalytic pocket by inhibitors could change the conformation of ACE2, thus stopingSARS-CoV-2 entry in host cells (Nutho et al., 2020;Towler et al., 2004).
In this study, we used N-acetylglucosamine (NAG) as the reference molecule. It is a naturally occurring amino sugar precursor for epithelial glycosaminoglycan synthesis and has been used to reduce chronic inflammation associated with osteoarthritis (McCarty et al., 2019). The basic science and clinical evidence support NAG as an anti-inflammatory and adjunctive treatment for COVID-19. It represented the scientific community's growing interest in nutraceuticals as potential COVID-19 treatments. Moreover, a similar nutraceutical, N-acetylcysteine (NAC), may have protective effects against COVID-19 complications due to its efficacy for other influenza viruses. Therefore in this regard, additional trials could shed light on the efficacy of NAG and other nutraceuticals for use as first-line or adjuvant therapies for COVID-19 (Hassan, 2021). Meanwhile, there are many therapies that have been explored and used to treat  Therefore, ACE2 is an attractive human target for coronavirus. Hence, to find out the potential inhibitor against the ACE2 receptor, we carried out virtual screening of phytochemicals of medicinal plants and reference molecule (NAG) against the ACE2. To gain a deeper insight into action, we further extended our study by ADMET, drug-likeness prediction, MD simulation, etc., which resulted in many potential phytochemicals that can further be tested in various experimental conditions to evaluate their efficacy against SARS-CoV-2.

Construction of phytochemical library of plants
Text mining analysis of medicinal plants used in the preparation of Chyawanprash was done by using server DLAD4U showed that these medicinal plants have anti-inflammatory, anti-diabetic, and potential antiviral activity. Hence, a library of 686 phytochemicals from 40 medicinal plants was constructed by searching the scientific literature (Supplementary Table 1). The 3 D structure of each phytochemical was downloaded from PubChem (https://pubchem.ncbi.nlm.nih.gov) in SDF format and then converted into PDB format using Open Babel (O'Boyle et al., 2011).

Protein preparation
The ACE2 receptor was selected as the target of COVID-19. The 3 D structure of the human ACE2 is covalently attached to its inhibitor. The X-ray crystal structure of ACE2 (PDB ID: 2AJF) was retrieved from the Protein Data Bank (http://www. rcsb.org/pdb/home/home.do). The structure of the protein was observed by PyMOL ( Figure 2) software.

Active site analysis
The active site analysis was done to calculate x, y, and z coordinates of a bounded ligand with receptor. The coordinates of the active site of the ACE2 receptor (PDB ID 2AJF) were generated by PyMOL tool. The active site of 2AJF Achain ligand NAG (1053) residues are Pro389, Met383, Pro559, Tyr515, His505, Pro273, Pro538, Gly537, His535, Cys530, Gln531, Cys530, Glu527 and Asn586, and these residues have been also described as catalytic residues. The active site residue Met389, Pro559, Tyr515, His505, Gln531, Glu527, Asn586 are a-Helix residue and Pro389, Pro273, both are b-turn residue while some active side residues are present in the region where disulfide bond formation occurs i.e., Pro534, Pro538, Gly537, His535, Cys530. The grid center for Figure 1. Finding potential inhibitor to inhibit ACE2-mediated SARS-CoV-2 infection. SARS-CoV uses the ACE2 receptor for cell entry has important implications for understanding SARS-CoV-2 transmissibility. SARS-CoV-2 lead to down regulation of the ACE2 receptor, but not ACE, through binding of the s-protein with ACE2. This leads to virus-host interaction, to viral entry and replication, as well as severe lung injury. blocking the surface ACE2 receptor by using ACE2 inhibitor which should slow viral entry into cells through competitively binding with SARS-CoV-2 and hence decrease viral spread as well as protecting the lung from injury through its unique enzymatic function.  Figure 3.

Molecular docking
Very first docking was conducted with ACE2 receptor and co-crystallized ligand to validate docking protocol using PyRx open-source software (GUI version 0.8 of AutoDock) in the active site of the receptor with help of AutoDock Vina (Trott & Olson, 2010). After that, virtual screening of the phytochemical library was conducted to select the best inhibitor for ACE2 receptor. Throughout the docking study, the ligand molecules were kept flexible while macromolecule as rigid. Finally, the binding energy table was extracted from the output files. The phytochemicals showed an excellent binding affinity with the receptor, were considered as best inhibitors (Verma et al., 2017). Visualization of 2 D interactions of the receptor-ligand complexes was carried out using Discovery Studio Visualizer and 3 D structure by PyMOL software. The RMSD calculation of docked complexes, i.e., reference molecule and other screened complex and co-crystalized molecule with receptor was done by using PyMOL software. PyMOL software calculate RMSD of docked complex and cocrystalized ligand by using the following command PyMOL > align dockedout, experimental, cycles ¼ 0, transform ¼ 0

Drug likeness prediction and ADMET analysis
Molecular properties, drug-likeness, and toxicity prediction of the screened compound are important steps in drug discovery. Therefore, all screened ligands were evaluated for their drug-like nature under Lipinski's five rules; 'RO5' (Lipinski, 2000) by DruLiTo open-source software. The drug, possessing good Absorption, Distribution, Metabolism, Excretion, and Toxicity (ADMET) parameters, could be approved easily by the FDA. Therefore, we have used the admetSAR server (http://lmmd.ecust.edu.cn:8000/) for predicting various parameters like Blood-Brain Barrier (BBB), Human Intestinal Absorption (HIA), Caco-2 cell permeability, P-GP (P-glycoprotein) substrate/inhibition, Cytochrome P450 metabolism, toxicity, carcinogenicity, Log S, and LD50 values.

Swiss target prediction
Swiss Target Prediction online server was used for predicting the percentage proportion activity of the selected phytoconstituents versus Chloroquine with known intracellular targets like kinases, nuclear receptors, transcription factors, phosphodiesterases, oxidoreductases, cytochrome P450, voltage gated-ion channels, hydrolases, phosphatases, G-protein coupled receptors and primary active transporters (Aniyery et al., 2015).

Bioactivity score
Compounds having druglike properties were further subjected to predict bioactivity by using Molinspiration which is online Cheminformatics software (Martin, 2005).

Pharmacophore study
Common pharmacophores of ligands responsible for biological activity against COVID-19 were analyzed using PharmaGist web servers. All screened phytochemicals along with reference molecule and reference drug Chloroquine were subjected for ligand-based pharmacophore studies.

Molecular dynamics simulation and MMPBSA calculation
The dynamics behaviors of transition at the atomistic level in the ACE2 of COVID-19 upon small binding molecules were investigated using MD simulations. MD simulations were performed using the CHARMM 36 force field embedded in the GROMACS 5.0.7 suite on the LINUX-based platform (Pronk et al., 2013). For the MD simulation, the CGenFF server was used to generate small molecules and protein topology files. Then all these systems were solvated by using the SPC water model in a cubic box. Eight Cl_ ions were added for the neutralization of all the systems and then energy minimization was carried out for removing the steric clashes of the system. Then the NVT and NPT simulations of 1 ns were carried out for maintaining the pressure (1 atm) and temperature (300 K) of the system. After that, the final MD production run of 100 ns was performed with an integration time step of 2.0 fs.
After the simulation, the Root means square deviation (RMSD), Root mean square fluctuation (RMSF), radius of gyration (Rg), and PCA (Padhi et al., 2021) were estimated by using GROMACS. Molecular Mechanics-combined with Poisson-Boltzmann (MMPBSA) method is used to evaluate binding free energies of small molecules with the target protein from the following equation: For each protein complex, the binding free energy was calculated from the last 10 ns simulation trajectories.

Molecular docking
The center of mass of active site prediction was done by PyMOL software. After that the docking protocol was used for doing the virtual screening. The binding energy of the reference compound (NAG) was À6.9 Kcal/mol. Virtual screening results of 20 phytochemicals showed a higher binding affinity in comparison to a reference compound. The ranges of the binding affinity of screened phytochemicals were À6.9 to À11.8 kcal/mol ( Figure 4). The docking score of screened phytochemicals with a receptor has been given in Table 1. The Reference compound (NAG) shows interaction with the several residues of 2AJF and forms three hydrogen bonds with His 535, Leu539, and Gly537 and six hydrophobic bonds with Lys534, Asn586, Glu527, Gln531, Cys530, and Pro538 and yields the binding energy is À6.9 kcal/mol ( Figure 5). The RMSD value between co-crystalized ligand (NAG) and docked ligand (NAG) was 0.537 and other screened complex Hinokinin (0.356), Isocolumbin (0.564), Tinocordioside (0.696) and Gmelanone (0.985), respectively. We calculated it by PyMol software. It showed that these complexes were aligned with reference molecule position and all four all docked complexes showed an acceptable range of RMSD. The acceptable range of RMSD is considered from 0 to 1.2 Angstrom.

Drug-likeness and ADMET prediction
Screened 20 phytochemicals were subjected to drug-likeness analysis in which only five phytochemicals were screened. All five screened, i.e., Isocolumbin, Tinocordioside, Gmelanone, Tinocordiside, and Hinokinin, 'drug-like' molecules have Log P value found with the range 2.409, 1.792, 0.548, 0.54, 2.011,  respectively. Predicted Log P values showed that screened phytochemicals could be absorbed in the body. All five phytochemicals have a molecular weight of 500 g/mol, the number of hydrogen bond acceptors 10, and the number of hydrogen bond donors 5. All the filters of drug-likeness profiles of the top five screened phytochemicals and reference compounds (NAG) have been compiled in Table 2. Five screened phytochemicals, as well as the reference compound (NAG), satisfied Lipinski's rule.
The ADMET properties of the top five screened phytochemicals were evaluated using the admetSAR database. Table 3 illustrates the relative ADMET profiles of the five phytochemicals compared to the reference (NAG). The Log S value refers to the solubility of the phytochemicals that ideally range between À6.5 and 0.5. All the phytochemicals are showing Log S values under this range. Among all the screened phytochemicals, Isocolumbin has the minimum Log S value (À4.3620), and Gmelanone has the maximum Log S value (À3.3933), which is similar to reference NAG (À3.4644). All compounds have above 30% Human Intestinal Absorption (HIA). If a compound with the HIA % is less than 30%, it is labeled as HIA-otherwise, it is labeled as HIAþ.
CaCO 2 and Blood-Brain Barrier permeability (BBB) can assess the permeability of the membrane. The CaCO 2 (colorectal carcinoma) permeability value for all the hit phytochemicals was comparable to the reference molecule, NAG (0.79). The computational BBB value corresponds to its entry into the central nervous system. The acceptable range of BBB values for an ideal drug candidate ranges between À3.0 and 1.2 (Nisha et al., 2016). All the phytochemicals have the BBB value under this range. Other distribution factors include renal organic cationic transporter and P-glycoprotein noninhibition. The major enzyme considered to examine ADME's metabolism parameter was Cytochrome P450 because of its role in Phase I drug metabolism. CYP450 is a group of enzymes that play a key part in drug and fatty metabolism (Guengerich, 2003). The hit phytochemicals exhibited acceptable results in toxicity analysis (Table 3). All hit phytochemicals cleared the Ames test except Tinocordiside. All screened phytochemicals were non-carcinogenic and no skin sensitization. Rat acute toxicity LD50 value of the reference molecule, all screened phytochemicals, is given in Table 3. Lethal Dose50 refers to the dose required to kill 50% of the population. These phytochemicals, viz., Hinokinin, Gmelanone, Isocolumbin, Tinocordioside, have 'no toxicity and no risk' except Tinocordiside. Thus, from the different ADMET values, it was observed that four screened compounds fulfill all the enlisted criteria as a drug in terms of the ADMET profile.

Bioactivity score prediction
Based on the Molinspiration portal, the bioactivity score of drug-like compounds are shown in Table 4. Compounds those have a bioactive score greater than 0 are said to be bioactive, score in between 0 and À0.5 shows moderate bioactivity and compounds having scored less than À0.5 are inactive. According to bioactivity score, four compounds pass the bioactive criterion i.e., they scored more than À0.5.

Swiss target prediction
Prediction of protein targets for selected phytoconstituents versus reference drug chloroquine was done using SwissADME. Figure 6 depicts the % bioactivity of selected phytoconstituents and reference drug choloroquine with respect to selected protein targets viz. enzymes, kinases, oxidoreductases, phosphatases, proteases, phosphodiesterases and lyases. Results showed that Isocolumbin displayed maximum activity towards AG protein-coupled receptor family (26.7%)and towards kinases (33.3%). Tinocordiosides was active with respect to phosphatase(33.3%) and other enzymes(20.0%), Hinokinin activity towards enzymes (26.7%) and AG protein-coupled receptor family (20.0%), Gamelanone showed activity towards AG protein coupled receptor family (26.7%) and other enzymes (20.0%) as well as electrochemical transport (20.0%), Chloroquine was active towards AG protein-coupled receptor family (60%) and enzymes (13.3%). Most of the phytoconstituents were found to be active towards several classes of soluble enzymes like kinases, lyases, oxidoreductases as well as membrane bound enzymes and transporter proteins. Interestingly, these phytochemicals showed potent binding to human ACE2 which is a membrane-bound enzyme and transporter protein that acts as the receptor for the spike S protein of the coronaviruses.

Pharmacophore study
To analyze the important features of phytochemicals and reference molecule and dug, we used the PharmaGist server. A set of structural features in molecules was recognized and found responsible for that molecule's biological activity in a receptor site. These structural features are known as pharmacophore features. The number of features and spatial feature set for each phytochemical from pharmacophore generation are summarized in Table 5. These features are useful for binding with receptors. As compared to reference screened phytochemicals also have many spatial features that are responsible for binding with receptors. These phytochemicals have aromatic, hydrophobic, donor, and acceptor properties which is necessary for potential anti-viral activity (Figure 7).

Molecular dynamics simulation
MD simulation is proven in silico methods for investigating the protein's real-time dynamics and conformational stability by binding the ligand. To know the conformational stability, structural stability, folding properties, and compactness of   protein-ligand combinations, MD simulations were conducted throughout 100 ns for five selected systems (Hinokinin, Isocolumbin, Tinocordioside, Gmelanone, and Reference). The average RMSD, RMSF, and Rg values are shown in Table 6. The ACE2 protein (black) and ACE2 with reference compound (NAG) (Red) showed the stable and constant range of RMSD between 0.25 and 0.30 nm, and similar results showed by ACE2-Hinokinin (Yellow), Isocolumbin (Green), and Gmelanone (Magenta), only slight changes showed in starting of the simulation. The Tinocordioside (blue) showed stable RMSD between 5 and 25 ns at RMSD range between 0.20 and 0.25 nm, after 25-55 ns, RMSD increases from 0.35 to 0.43 nm. The Tinocordioside showed higher but stable RMSD between 0.31 and 0.45 nm throughout the simulation. Tinocordioside binding affects the RMSD of protein, but it is stable till the end of the simulation. The differences of RMSD with Hinokinin, Isocolumbin, and Gmelanone complex in protein RMSD suggest that ACE2 has conformational changes in Tinocordioside. In contrast, no significant change has shown Hinokinin, Isocolumbin, and Gmelanone after 60 ns simulation (Figure 9(a)).
Additionally, the stability of the docked complex was calculated by measuring the residue flexibility. Analysis of the Root means square fluctuation (RMSF) graph revealed no significant fluctuations in the active site region. In addition, most protein residues were stable, with RMSF values below 0.3 nm. The reference complex was showed fluctuation of 0.35 and 0.38 nm with residue Gln300 and Pro438. ACE2-Tinocordioside was showed fluctuation of 0.38, 0.43, and 0.39 nm in residue Glu56, Pro138, and Lys340. The ACE2-Gmelanone complex was showed fluctuation in the range 0.40 nm with Pro138 and 0.55 nm with Lys340. The ACE2-Hinokinin complex was showed fluctuation of 0.30 nm with Leu100. Results revealed that these residues could play a crucial part in stabilizing the complex (Figure 9(b)).
Rg the compactness of the system with the time, Rg was calculated, in which higher Rg values showed less compactness, while low Rg values explain high compactness. As evident from Figure 9(c), the simulation Rg values of the native protein and all protein-ligand complexes were reported as 2.50-2.55 nm. Rg results showed that the binding of these four molecules does not induce structural changes. The Rg values of all four protein-ligand complexes support their condensed architecture as well as size. The ACE2-Tinocordioside complex was showed fluctuation 5-70 ns; the Rg value is 2.55-2.65 nm. Besides it, after 70 ns, the Rg value of the ACE2-Tinocordioside complex is more stable than other complexes (Figure 9(c)).

Principal component analysis (PCA)
PCA was carried out to investigate the significant motions during ligand binding. In this study, the diagonalization of the matrix was used to calculate the eigenvectors and  ACE2-reference complex 0.25 ± 0.02 0.13 ± 0.03 2.16 ± 0.02 3.
ACE2-Hinokinin complex 0.24 ± 0.02 0.14 ± 0.04 2.16 ± 0.02 eigenvalues. For this study, the first 20 eigenvectors were selected to calculate concerted motions. Figure 10(b) represents the eigenvalues that were obtained from the diagonalization of the covariance matrix of atomic fluctuations in decreasing order versus the corresponding eigenvector for ACE2-Reference complex (NAG), ACE2-Isocolumbin, ACE2-Tinocordioside, ACE2-Hinokinin, ACE2-Gmelanone complex. The 2 D projection plot generation in PCA is another way to achieve the dynamics of complexes. Figure 10(a) shows the 2 D projection of the trajectories in the phase space for the first two principal components, PC1 and PC2 for ACE2reference, ACE2-Isocolumbin, ACE2-Tinocordioside, ACE2-Hinokinin, and ACE2-Gmelanone complexes, respectively. The complex which occupied less phase space, showed a stable cluster while the complex that occupied more space showed a less stable cluster. From the figure, it can be concluded that the ACE2-Isocolumbin, ACE2-Tinocordioside, and ACE2-Gmelanone were highly stable as they occupied less space in the phase space, and the cluster was well defined as compared to ACE2-reference, and ACE2-Hinokinin complexes. It was observed that out of the 20 eigenvectors, the first ten eigenvectors accounted for 86.29% for ACE2-reference, 87.23% for ACE2-Isocolumbin, 87.83% of total motions for ACE2-Tinocordioside, 82.25% for ACE2-Hinokinin, and 86.95% for ACE2-Gmelanone complexes (Figure 10(b)). All the studied complexes showed very fewer motions as compared with the reference compound. Therefore, from the PCA analysis, we concluded that all complexes showed very fewer motions and form a stable complex with ACE2-spike protein.
The 2 D projection plot generation in PCA is another way to achieve the dynamics of complexes.
Gibb's energy plot for PC1 and PC2 was also calculated and is shown in Figure 11. The plot shows Gibbs energy value ranging from 0 to 13.6 for ACE2-reference, Gibbs energy is 0-10.1 kJ/mol for ACE2-Gmelanone, Gibbs energy is 0-9.97 kJ/mol for ACE2-Isocolumbin, forACE2-Hinokinin Gibbs energy is 0-13.7 kJ/mol, and for ACE2-Tinocordioside Gibbs energy is 0-10.7 kJ/mol, respectively. Thus, screened compounds had lower energy as compared to the reference withACE2, which indicates that these complexes follow energetically more favorable transitions from one conformation to another as compared to the reference.
To determine the Binding pose of protein-ligand complex the snapshot of all four compounds and references were extracted at 100 ns time intervals. The result reveals that the selected four compounds and reference most of the time stayed in the binding site of the protein during the 100 ns simulation period (Figure 12).

MMPBSA calculation
Although the free binding energy is calculated from the MM/ PBSA method, it differs from the docking calculations. They follow the same energetic order. ACE2-Tinocordioside complex, ACE2-Gmelanone complex, and ACE2-Isocolumbin complex, ACE2-Hinokinin complex present the same binding energy compared to reference with similar van der Waals and a polar solvation (SASA) energies (Table 7). ACE2-Hinokinincomplex (À4.739 KJ/mol) showed higher free binding energy. The Protein ligands complexes' simulations show that the ligands remain bounded in the RBD region along with the simulation where the binding energies obtained from MM/PBSA calculations are shown in Table 7.

Discussion
COVID-19 started in Wuhan, and it has rapidly spread to almost all over the world. The novel coronavirus uses ACE2 to infect human. ACE2 took centre stage in the COVID-19 outbreak as the key receptor for the spike glycoprotein of SARS-CoV-2, as demonstrated in multiple structural and biochemical interaction studies. ACE2 is expressed more abundantly on polarised epithelial cells' apical surface, which shows that well-differentiated cells support viral replication from the apical surface (Jia et al., 2005). Thus, SARS-CoV preferentially infects well-differentiated ciliated epithelial cells expressing ACE2 (Hofmann et al., 2005). More death cases of COVID-19 occurred in older people, possibly due to a weak immune system that permits faster viral infection (Li et al., 2020;Wang et al., 2020). The immune response is considered very important for the control and resolution of SARS-CoV-2 infection. Transmissibility of SARS-CoV-2 is much greater than that of SARS-CoV, suggesting that SARS-CoV-2 might For the coronaviruses, there are currently no specific therapies available. Symptomatic and supportive treatment are the mainstay for these viruses. The ongoing COVID-19 pandemic has posed serious threats to global health (Bloom & Cadarette, 2019) for which there is a need to discover and evaluate novel therapeutic agents having antiviral efficacy against SARS-CoV-2. Unfortunately, there has been no significant improvement in COVID-19 disease management till date. While there have been several attempts to study and improve diagnostics, research is lagging as far as therapeutics for SARS-CoV-2 and its recent mutant strains are concerned (Hajivalili et al., 2020). In this context, a number of traditional drugs viz. Chloroquinine, Hydro Chloroquinine, remdesivir, etc., have been repurposed for use as therapeutic antiviral agents and have been tested in vitro with some curative effects to counter SARS-CoV-2. Though the clinical efficacy of these drugs has shown promise, but the problem of associated drug toxicity looms large (Khan & Al-Balushi, 2021). In the present scenario, with no safe and effective therapeutic and vaccine agents available to treat COVID-19, The best option is to use Chyawanprash to prevent COVID-19, it is regarded as a very effective immunity booster and a preventer of day-to-day infections & allergies such as common cold and cough, etc (Gr€ uber et al., 2008). In preparation of Chyawanprash Phyllanthus emblica is used in large amount. The extract of Phyllanthus emblica extract that contains Emblicanin A and Punigluconin molecules are also reported for antiviral activities against different groups of viruses and as an immunomodulator (Sharma et al., 2020). In this study, we screened antiviral phytochemicals from 40 medicinal plants of Chyawanprash against ACE2, a receptor of COVID-19. To find out the inhibitor for the ACE2 receptor, we made a library of 686 phytochemicals of 40 medicinal plants of Chyawanprash. Only 20 phytochemicals are screened out by molecular docking. After that, the drug-likeness prediction gave only five phytochemicals, Hinokinin, Gmelanone, Isocolumbin, and Tinocordioside, and the toxicity prediction of screened five phytochemicals was also done. The key residues of inhibitor are His535, Leu539, Gly537, Lys534, Glu527, Gln531, Cys530, Pro538, and Asn586, and these residues have been also described as catalytic residues. The functional group of N-Acetylglucosamine which is co-crystalized ligand in ACE2 receptor is amide derivative. It is a secondary amide between glucosamine and acetic acid. In our results. The screened Hinokinin, Isocolumbin, Tinocordioside and Gmelanone showed Furan group. Several research groups studied the antiviral properties of hinokinin against human hepatitis B virus (HBV) (Huang et al., 2003), human immunodeficiency virus (HIV), SARS-virus (SARS-CoV) (Wen et al., 2007), and in all cases it showed significant antiviral activity.
Only four complexes showed no toxicity and no carcinogenic, while Tinocordiside showed AMES toxicity. To analyze, the binding site of phytochemicals 2 D and 3 D structure was generated through ligplot and PyMOL software. To know the stability of complexes, these four compounds were selected for MD simulation. The free binding energy obtained from MM/PBSA calculations were ACE2-reference complex (7.327 ± 92.719), ACE2-Tinocordioside complex (À0.679 ± 80.244), ACE2-Gmelanone complex (1.735 ± 50.130), ACE2-Isocolumbin complex (À85.198 ± 93.690), and ACE2-Hinokinin complex (À1.693 ± 61.407 kJ/mol), respectively. From all complexes, Isocloumbin showed good binding energy in comparison to the reference molecule. These four phytochemicals belong to three medicinal plants, viz. Phyllanthus amarus, Gmelina arborea, and Tinospora cordifolia. The maximum number of ACE2 inhibitors was screened from Tinospora cordifolia, which was also used for fever and boosted the immune system managing diabetes and made our respiratory system stronger (Singh et al., 2003). Phyllanthus amarus and Gmelina arborea have anti-inflammatory and antiviral properties. These five phytochemicals followed the drug-likeness rule, and they have no toxicity. We suggest that these screened phytochemicals may be a potential drug candidates for inhibition of ACE2 and Spike protein interaction.

Conclusion
ACE2 is a promising target against the current global threat of COVID-19 pandemic, caused by SARS-CoV-2. In view of the same, natural compounds may hold a prominent role in therapeutics discovery to mitigate the viral infection. In the present work, four promising phytochemicals as ACE2 inhibitors have been identified by using intensive in silico screening methods, namely, Hinokinin, Gmelanone, Isocolumbin, and Tinocordioside, from the plants Vitis vinifera, Gmelina arborea, and Tinospora cordifolia respectively. All the screened hits showed acceptable range of drug-likeness and ADMET profile with excellent binding profile with ACE2 confirmed by molecular docking and MD simulation studies up to 100 ns. The compounds Tinocordioside and Isocolumbin were reported from the plant Tinospora cordifolia, commonly known as 'Guduchi' or 'Giloy' or 'Amrita'. It is a well-known plant in the Indian System of Medicine (Ayurveda) as 'Rasayana' that possesses anti-inflammatory, anti-arthritic, anti-allergic, anti-malarial, anti-diabetic, hepatoprotective, immunostimulatory, hyperlipidemic and aphrodisiac properties. The compound Gmelanone was reported from the plant Gmelina arborea and Hinokinin was from Phyllanthus amarus also known as 'Bhumyamalaki'. All of the above-mentioned plants are well documented ayurvedic plants with excellent medicinal properties. The compounds Gmelanone and Hinokinin are already commercially available in market, whereas Tinocordioside and Isocolumbin may be synthesized commercially for further for future biological validation tests and therapeutic use of the compounds against SARS-CoV-2. However, all compounds may be also directly isolated from their concerned plants cultivated commercially that will promote value chain addition to these ancient ayurvedic plants. Therefore, a very promising future direction in the research of novel anti-coronavirus agents could be represented by the lead optimization of such a chemical scaffold. Further in vitro, in vivo, pre-clinical, and clinical phases studies are necessary in order to verify the antiviral activity of the top rating molecules against SARS-CoV-2 and hopefully their ability to alleviate the emergence as well as surmount drug-resistant viral strains.