key: cord-1018659-mrjjtwv8
authors: Alamri, Mubarak A.; Altharawi, Ali; Alabbas, Alhumaidi B.; Alossaimi, Manal A.; Alqahtani, Safar M.
title: Structure-based virtual screening and molecular dynamics of phytochemicals derived from Saudi medicinal plants to identify potential COVID-19 therapeutics
date: 2020-08-09
journal: Arabian journal of chemistry
DOI: 10.1016/j.arabjc.2020.08.004
sha: e4da91124f6ce9c5d4afbcc6c7caedc45eae3676
doc_id: 1018659
cord_uid: mrjjtwv8

Abstract Coronavirus disease 2019 (COVID-19) has affected almost every country in the world by causing a global pandemic with a high mortality rate. Lack of an effective vaccine and/or antiviral drugs against SARS-CoV-2, the causative agent, has severely hampered the response to this novel coronavirus. Natural products have long been used in traditional medicines to treat various diseases, and purified phytochemicals from medicinal plants provide a valuable scaffold for the discovery of new drug leads. In the present study, we performed a computational screening of an in-house database composed of ∼1000 phytochemicals derived from traditional Saudi medicinal plants with recognised antiviral activity. Structure-based virtual screening was carried out against three druggable SARS-CoV-2 targets, viral RNA-dependent RNA polymerase (RdRp), 3-chymotrypsin-like cysteine protease (3CLpro) and papain like protease (PLpro) to identify putative inhibitors that could facilitate the development of potential anti-COVID-19 drug candidates. Computational analyses identified three compounds inhibiting each target, with binding affinity scores ranging from-9.9 to -6.5 kcal/mol. Among these, luteolin 7-rutinoside, chrysophanol 8-(6-galloylglucoside) and kaempferol 7-(6’’-galloylglucoside) bound efficiently to RdRp, while chrysophanol 8-(6-galloylglucoside), 3,4,5-tri-O-galloylquinic acid and mulberrofuran G interacted strongly with 3CLpro, and withanolide A, isocodonocarpine and calonysterone bound tightly to PLpro. These potential drug candidates will be subjected to further in vitro and in vivo studies and may assist the development of effective anti-COVID-19 drugs.

The global threat of the 2019 novel coronavirus disease (now officially named COVID-19 and the causative pathogen is SARS-CoV-2) is rapidly escalating with unprecedented international health and economic burden in the recent history. The whole-genome of SARS-CoV-2 had been sequenced and revealed that SARS-CoV-2 pathogen is the fifth strain of βcoronaviruses, which include OC43, HKU1, SARS-CoV-1 and Middle East respiratory syndrome coronavirus (MERS-CoV) . After the entry into the host cell through strong binding of β-coronaviruses' protein spikes with the human angiotensinconverting enzyme 2 (ACE2) receptor , β-coronaviruses generate large polyproteins (PP1a and PP1ab) upon genome translation of open reading frame (ORF) 1a and

ORF1ab by the host cell machinery. These polyproteins, also known as replicase polyproteins, are proteolytically cleaved by essential cysteine proteases encoded by the virus, explicitly 3chymotrypsin-like protease (3CL pro ; sometimes called main protease (M pro )) and papain-like proteases (PL pro ), to release 16 non-structural proteins (nsps) (Thiel et al., 2003; Ziebuhr, 2004) . Among those nsps is the RNA-dependent RNA polymerase (RdRp or sometimes referred to as nsp12). RdRp catalyses the synthesis of a complementary RNA strand using the viral RNA and hence, it plays an essential role in directing the replication and transcription of SARS-CoV-2 genome Lung et al., 2020; Zumla et al., 2016) . The proteolysis of PP1a and PP1ab by 3CL pro occurs at 11 distinct sites and generates various nsps that are important for the viral replication (Anand et al., 2003) . Therefore, 3CL pro is crucial for the virus particle replication and represent a valid target for the identification of coronavirus inhibitors (Needle et al., 2015) . PL pro is another crucial cysteine protease, an enzyme that cleaves N-terminus of the replicase polyprotein to release several nsps, among them the nsp3, in which PL pro is encoded and is implicated not only in the viral replication but also in suppressing the host innate immune response Lei et al., 2018; Yuan et al., 2015) . In addition, PL pro possesses a nucleic acid-binding domain (NAB) with a nucleic acid chaperon function and is essential in the virus replication correction. Furthermore, PL pro has posttranslational modification activities, deubiquitination and deISGylating, on the host proteins, such as interferon regulatory factor 3(IRF3), which consequently lead to the removal of ubiquitin and human interferon-stimulated gene 15 (ISG15), the main signalling elements of the antiviral innate immune response (Báez-Santos et al., 2015; Chen et al., 2020; Lei et al., 2018; Liu et al., 2020) . The indispensable role of the RdRp, 3CL pro and PL pro in orchestrating the RNA genome replication/transcription cycles of SARS-CoV-2 make them potential targets for designing antiviral candidates against COVID-19 (Báez-Santos et al., 2015; Ramajayam et al., 2011; Ren et al., 2013) .

For ages, phytochemicals have been found to be a fruitful source of molecules with diverse therapeutic potentials and still are considered a valuable resource for the discovery of novel drug leads In other word, herbal medicines and purified phytochemicals can be used to develop more efficient drugs based on the structure of natural compounds (Mani et al., 2020) . Given the time of its emergence, few studies related to the development of naturally-derived inhibitors of three main druggable targets of SARS-CoV-2 (RdRp, 3CL pro and PL pro ) have already been reported using computer modelling for screening purposes. For instance, in silico or biological screening of a series of biologically active natural compounds have been demonstrated to directly inhibit these important proteins in pervious HCoVs such as SARS-CoV, MERS-CoV (Park et al., 2012; Shen et al., 2019) and the novel SARS-CoV-2 (Aanouz et al., 2020; Ul Qamar et al., 2020; Zhang et al., 2020) . (Lung et al., 2020) virtually screened 83 compounds found in Chinese traditional medicines with potential activity against SARS-CoV-2 and identified theaflavin as a potential inhibitor of the RNA-dependent RNA polymerase. Likewise, screened 115 compounds that are commonly used in treating viral respiratory infection in China and highlighted 13 compounds for further studies to prove their activities against SARS-CoV-2. A virtual screening and molecular modelling study of marine natural products have also been investigated and proposed seventeen molecules with potent inhibitory action against 3CL pro (Gentile et al., 2020) . In the present study, we conducted a computational study on a library of phytochemicals from Saudi medicinal plants with potential antiviral activities (Aati et al., 2019; Arbab et al., 2017) to identify potent inhibitors of RdRp, 3CL pro and PL pro , which eventually can be employed as novel lead molecules to accelerate the pace of designing anti-COVID-19 candidates.

The X-ray crystal structures of three SARS-CoV-2 proteins including the RNAdependent RNA polymerase (RdRp) with 2.50 Å resolution (accession number 7BV2) (Yin et al., 2020) , 3-chymotrypsin-like cysteine protease (3CL pro ) with 2.1 Å resolution (accession number 6W63) (Mesecar, 2020) and papain like protease (PL pro ) with 2.7 Å resolution (accession number 6W9C) (Osipiuk et al., 2020) were obtained from the Protein Data Bank.

Discovery studio program (Studio, 2008) was used to prepare the proteins for virtual screening and molecular docking by removing co-crystalized ligands and water molecules and to prepare the proteins in monomer form in PDB formats.

A library containing ~1000 plant-derived compounds obtained from 60 traditional Saudi medicinal plants with antiviral activity were designed. The phytochemicals were collected by comprehensive literature study. The 3D structures of compounds were obtained either from PubChem database (available online: https://pubchem.ncbi.nlm.nih.gov/ ) or were drown manually using ChemDraw Ultra 7.0 and saved in one SDF file.

The virtual screening of phytochemical database against the structure of target proteins was performed individually using Autodock vina in PyRx 8.0 virtual screening tool (Dallakyan et al., 2015) . At first, compounds were imported into OpenBabel program (Babel, 2011) implemented in PyRx for energy minimization using MMFF94 force field and for converting compounds into Autodock PDBQT format. The 3D grid box parameters were set to cover the active site cavity within each protein. The top three compounds with highest binding energy scores against each viral protein were selected as hit compounds for further assessment.

The molecular docking of candidate phytochemicals against their viral protein's targets were carried out using Autodock vina 1.1.2 program (Trott et al., 2010) . Autodock Tools 1.5.6

program (Huey et al., 2008) was employed to prepare the PDB structures of proteins for docking by adding polar hydrogen and to convert proteins and ligands PDB files into Autodock PDBQT format. The grid box parameters for RdRp were set to size 24Å x 22Å x 28Å (x, y and z) and centre 113.78Å x 118.474Å x 132.008Å (x, y and z). For 3CL pro , the grid box parameters were set to size 18Å x 18Å x 18Å (x, y and z) and centre -20.318Å x 19.141Å x -28.343Å (x, y and z). As for PL pro , the grid box parameters were set to size 21Å x 22Å x 22Å (x, y and z) and centre -27.423Å x 30.003 Å x 27.531Å (x, y and z). The 2D ligand-protein interaction diagrams were generated by PoseView tool (available online: https://proteins.plus/ ). PyMOL (DeLano, 2002) was used to visualize and analyse the docking results.

All atom molecular dynamic (MD) simulation of ligand-protein complexes were performed with GROMACS 2018 software OPLS-AA/L force field. The high-resolution Xray structures of proteins (PDB IDs: 6M71, 6W63 and 6W9C for RdRp, 3CL pro and PL pro , respectively) were used as initial structures for the MD simulation. The protein structures were further optimized for simulation by DockPerp tool in UCSF Chimera (Pettersen et al., 2004) .

The top-predicted docking pose of each ligand with the highest binding score was used as a starting point for the MD simulation. The ligands were parametrized using SwissParam webserver (Zoete et al., 2011) . All systems were simulated for 20 ns using previously reported protocol (Alamri, 2020) . Basic MD simulation parameters including real-mean-squaredeviations (RMSD), real-mean-square-fluctuation (RMSF), radius of gyration (Needle et al., 2015) , number of hydrogen bonds (H-bonds) were analysed for each complex.

Initially, an in-house phytochemical database composed of ~1000 bioactive compounds derived from 60 traditional Saudi medicinal plants was designed. The extracts of the selected plants were reported to possess antiviral activity against a wide-range of viruses such as HIV, HCV and HCB at various concentrations (Aati et al., 2019; Arbab et al., 2017) . The main constituents of each plant were identified by comprehensive literature study. This approach was applied to build a focused phytochemicals library with recognised antiviral activity, so it can be further screened in search for potential anti-COVID19 agents. The tested library composed of phytochemicals that represent major classes of natural products as shown in (Fig.   1) . The natural occurrence and examples of the top five chemical classes are presented in (Table S1 ).

A high-throughput virtual screening of a diverse plant-derived compounds from Saudi medicinal plants, followed by molecular docking and all atom 20ns MD simulation studies were performed. The screening was conducted against three well-established drug discovery targets including SARS-CoV-2 RdRp, 3CL pro and PL pro enzymes.

First, we screened the designed database against SARS-CoV-2 RdRp. The RdRp is a conserved enzyme in coronaviruses and essential for their replication and transcription of their genomes Lung et al., 2020; Zumla et al., 2016) . RdRp in its active form tends to form a heterotrimeric complex with nsp7 and nsp8 for maximum activity . The virtual screening of in-house phytochemical database was performed against the recently solved X-ray structure of the nsp12 protein. The analysis resulted in the identification of three compounds, namely chrysophanol 8-(6-galloylglucoside), luteolin 7-rutinoside and kaempferol 7-(6''-galloylglucoside), which have been reported to be isolated form Rumex dentatus (Mishra et al., 2018) , Marrubium vulgare (Neamah et al., 2009) and Acacia species (El-Toumy et al., 2018) , respectively. The natural source and docking scores of these compounds are presented in (Table 1) .

In order to gain more insight into the mechanism of interaction of these compounds with SARS-CoV-2 RdRp, a molecular docking was conducted using Autodock vina. Initially, Remdesivir which is the only approved drug to treat patients with severe COVID-19 was used to validate the docking approach (FDA, 12 April 2020). Remdesivir is a nucleotide analogue prodrug of (GS-441524) that converted into the active triphosphate form inside the cell.

Recently, the structure of RdRp in complex with Remdesivir was solved (PDB: 7BV2).

Remdesivir was found to interact covalently in its monophosphate form with the primer strand at the +1 position within the active site of SARS-CoV-2 RdRp (Yin et al., 2020) . Fig. 2a showed that the redocked Remdesivir monophosphate pose binds to the active site within RdRp as same as the co-crystalized ligand with RMSD= 2.969 Å. This result validates the ability of our docking protocol to predict the bioactive conformation of inhibitors. In addition, Remdesivir monophosphate exhibited lower binding free energy (-7.1 kcal/mol) in comparison to candidate compounds. This implies a strong possibility for these ligands to bind RdRp with higher affinity ( Table 1) . The molecular docking results of identified ligands illustrated that all compounds adapted binding modes similar to Remdesivir monophosphate within the activesite of RdRp ( Fig. 2A) . The chrysophanol 8-(6-galloylglucoside) was found to interact with the active site through six H-bonds (Tyr619, Asn691, Lys621, Asp623, Asp760 and Asp761) and one hydrophobic interaction (Pro621) (Fig. 2b) . In the other hand, luteolin 7-rutinoside was involved in nine H-bonds with the active site (Arg553, Lys545, Arg555, Asp623, Asn691, Thr687, Asp760 and Ser759) (Fig. 2c). Fig. 2d , showed that kaempferol 7-(6''galloylglucoside) formed seven H-bonds with RdRp (Arg553, Thr556, Tyr619, Trp617, Lys621, Asp760 and Asp761). Notably, the catalytic site of RdRp is composed of seven conserved motifs from A to G. Similar to Remdesivir, candidate compounds interact via Hbonds with several residues including the conserved Ser759-Asp60-Asp761 sequence within motif C of the RdRp polymerase domain (Fig. 2B) . These results suggested that the selected hits might act as non-nucleotide anti-polymerase agents.

In order to assess RdRp association and to determine the dynamic stability of screened compounds, all atom MD simulation at 20 ns were carried out by GROMACS. Analysis of RMSD from the starting structure revealed that most of complexes reach equilibrium within 20 ns (Fig. 3A) . This points to the high stability of protein-ligand complexes. Some fluctuation was observed within the first 10 ns in all systems. For the structure of RdRp bound to Remdesivir monophosphate and Chrysophanol8-(6-galloylglucoside), the RMSD values stabilized around ~0.32 nm. On the other hand, the RMSD values for the RdRp complex with Luteolin 7-rutinoside and Kaempferol 7-(6''-galloylglucoside), were equilibrated at around ~0.24 and ~0.20 nm, respectively. These results suggested that the binding of Remdesiver monophosphate and Chrysophanol8-(6-galloylglucoside) to RdRp may induce conformational changes (Fig. 3A) . In consistent with this, the analysis of RMSF vs RdRp residue number showed that Remdesivir monophosphate and Chrysophanol8-(6-galloylglucoside) complexes showed higher oscillations in backbone residues in comparison to Luteolin 7-rutinoside and Kaempferol 7-(6''-galloylglucoside) systems (Fig. 3B) .

The RMSF values were high in some residues located in the flexible loop as well as alpha helices regions such as Thr261, Asp336, Glu431and Asp824. Radius of gyrate (Rg) is another parameter to assess the compactness changes of a ligand-protein complex. Remdesivir monophosphate displayed a high Rg fluctuation in comparison to other compounds. This is consistent with the docking results of Remdesiver monophosphate that showed the lowest binding free energy. In contrast, no significant changes were observed in Rg values of Chrysophanol8-(6-galloylglucoside) and Kaempferol 7-(6''-galloylglucoside) indicating sustained compactness of both systems throughout the simulation (Fig. 3C) . The Luteolin 7rutinoside complex was the most stable and compact system with Rg value of ~2.94 nm. Next, the formation and stability of H-bonds were investigated over the simulation time (Fig. 3D) .

The H-bond properties are essential parameter in drug design due to their role in drug specificity, metabolism and absorption (Bitencourt-Ferreira et al., 2019) . The results illustrated that Remdesiver monophosphate, Chrysophanol8-(6-galloylglucoside), Luteolin 7-rutinoside and Kaempferol 7-(6''-galloylglucoside) could form up to 5, 10, 11 and 10 H-bonds, respectively. Therefore, the stability of complexes was maintained by H-bonds formation with active site residues. According to the docking and MD simulation analyses, Luteolin 7rutinoside showed the best affinity towards RdRp in comparison to other compounds. It showed a high docking score (-9.8 kcal/mol) and was able to form stable H-bonds with the essential Asp760 and Asp761 of SDD sequence of polymerase domain. Moreover, the results from RMSD, RMSF and Rg indicated that the Luteolin 7-rutinoside complex was stable during the simulation period. Therefore, it can be concluded that Luteolin 7-rutinoside is the most preferable compound for the inhibition of RdRp for further in vitro and in vivo studies.

The 3CL pro , also known as the main protease (M pro ) is a cysteine protease. Its proteolytic activity is mediated by His41-Cys148 catalytic dyad (Anand et al., 2003; Needle et al., 2015) .

The structure based virtual screening of our phytochemical database led to identification of three compounds namely, 3,4,5-Tri-O-galloylquinic acid, chrysophanol 8-(6-galloylglucoside) and iso-mulbel-rochromene with docking scores of -9.7, -9.7-and -9.6 kcal/mol, respectively ( Table 2) . Interestingly, Chrysophanol 8-(6-galloylglucoside) was also identified as a potential RdRp inhibitor indicating that it might have a dual inhibitory mechanism.

To predict these inhibitors conformation and orientation inside the active site of 3CLpro, a molecular docking was carried out. All selected compounds were found to fit into the active site with similar binding mode (Fig. 4A) . The co-crystalized non-covalent ligand, X77, was re-docked to assess the capability of our docking protocol to re-produce the bioactive conformation. As shown in Fig. 4a , the docked pose adapted similar binding mode as the cocrystalized ligand with RMSD =0.789Å indicating the robustness of our docking procedure.

3,4,5-Tri-O-galloylquinic acid binds with the active site of 3CLpro through several H-bonds (Met49, Leu141, Ser144, Glu166, His163 and Asp187) (Fig. 4b) . The interaction between chrysophanol 8-(6-galloylglucoside) and 3CLpro was essentially mediated through H-bonds with four residues (Glu166, Arg188, Thr190 and Gln192) (Fig. 4c) . Iso-mulbel-rochromene formed six H-bonds with the active site of 3CLpro (Thr25, Leu141, Ser144, His163, Arg188 and Gln189). These compounds block most of the essential residues involved in substrate recognition through H-bonds interaction indicating that these compounds may have potential to inhibit SARS-CoV-2 3CL pro activity.

Next, we characterized the stability of dynamic behavior and binding conformation of 3CL pro in complex with co-crystalized ligand, X77 as well as the identified compounds by 20 ns all-atom MD simulations. Fig. 5A illustrated the plot of RMSD (nm) vs. time (ns) for apo-3CL pro and three 3CL pro -ligand complexes (3,4,5-Tri-O-galloylquinic acid, chrysophanol8-(6-galloylglucoside) and iso-mulbel-rochromene). All complexes attained the equilibration state during the 20ns MD simulation period. The 3CLpro-X77 complex showed higher RMSD value as compared to identified ligand complexes during the last 10 ns, whereas 3,4,5-Tri-Ogalloylquinic acid showed the least RMSD value, i.e., a greater dynamic stability compared to the other compounds. The RMSF values showed normal fluctuation with all ligand-protein complexes with respect to the RMSF value of co-crystalized ligand (Fig. 5B) . The Rg profiles were quite similar for all ligand-complexes as compared to X77-3CL pro complex. Thus, protein-ligands complexes were stable complexes and well folded (Fig. 5C ).

The number of H-bonds for ligand-protein complexes was calculated over 20ns simulation time. Fig. 5D showed that 3,4,5-Tri-O-galloylquinic acid, chrysophanol8-(6galloylglucoside) and iso-mulbel-rochromene had an average of eight, eleven and seven Hbonds, respectively, whereas an average of four H-bonds are formed in the X77-3CL pro complex. This explains the high binding energy scores of identified ligands compared to cocrystallized ligand, X77. These findings offer an extensive opportunity for these natural compounds, especially 3,4,5-Tri-O-galloylquinic acid, to contradict 3CL pro for the treatment of COVID-19.

The PL pro is another key cysteine protease enzyme for coronaviruses life cycle. In addition to its function in virus replication, it is involved in facilitating the virus entry by antagonizing the host cell immunity (Báez-Santos et al., 2015; Chen et al., 2020; Lei et al., 2018; Liu et al., 2020) . The virtual screening against PL pro revealed that withanolide A, isocodonocarpine and calonysterone exhibited the highest binding affinity scores of -7.4, -7.0 and -6.9 kcal/mol, respectively (Table 3) .

Next, we performed a molecular docking to determine the molecular interactions of these inhibitors with PL pro . At first, the docking protocol was validated by docking a known SARS-CoV inhibitor; GRL0617 (PDB ID: 3E9S) into the same site within SRS-CoV-2 PL pro (PDB ID: 6W9C) (Ratia et al., 2008) . This site is a conserved in PL pro for both SARS viruses.

It is located in a proximity to the catalytic site which is composed of catalytic triad residues (Cys112-His273-Asp287) of PL pro (Fig. 6A) . Notably, the initial virtual screening was carried out against this pocket. Fig. 6A showed that GRL0617 as well as all identified ligands fit into the binding pocket with similar binding modes. A closer view into the binding of GRL0617 to PL pro of SRAS-CoV-2 showed that it adapted a similar binding mode and conformational arrangement for SARS-CoV (PDB ID: 3E9S), which demonstrated the validity of docking approach (Fig. 6a) . Withanolide A interacted with the active site of PL pro through three Hbonds with Lys157, Asp164 and Glu167 (Fig. 6b) . In the other hand, the cyclic compound, isocodonocarpine formed hydrophobic interactions with Tyr264 and Tyr268 and π-π stacking with Tyr264 (Fig. 6c) . The calonysterone's binding was found to involve in three H-bonds with Glu161, Asp164 and Tyr246 and one hydrophobic interaction with Pro248 (Fig. 6d) .

Similarly, to ensure the stability of ligands within the active site of PL pro , 20 ns all-atom MD simulation using GROMAS 2018 was performed. The RMSD results of withanolide A-PL pro , isocodonocarpine-PL pro and calonysterone-PL pro complexes reached a stable state throughout the simulation period. The fluctuation in the RMSD values of withanolide A-and isocodonocarpine-PL pro complexes were less compared to the apo-protein and calonysterone-PL pro complex (Fig. 7A) . The RMSF values for isocodonocarpine showed less RMS fluctuation than other complexes, with respect to the RMSF values of apo-protein (Fig. 7B) .

illustrated that all ligand-protein complexes showed similar Rg behaviour, with respect to the apo-protein suggesting stable ligand-protein complexes in general. The H-bonds were also measured for each complex (Fig. 7D) . Withanolide A, isocodonocarpine and calonysterone showed an average of two, three and five H-bonds. Although, isocodonocarpine did not show any H-bonds by molecular docking, three H-bonds were predicted by MD simulation. In addition, the number of H-bonds formed by withanolide A were almost stable, while the number of H-bonds formed by calonysterone was decreased over the MD simulation.

This could explain the low docking score of calonysterone. Overall, Withanolide A showed the best results with respect to docking score as well as the MD simulation analyses suggesting that it might be a potential PL pro inhibitor candidate to combat COVID-19 infection.

In the presented study, we employed computational screening approach for the discovery of potential anti-COVID-19 against the novel coronavirus SARS-CoV-2 infection.

A total of ~1000 phytochemicals derived from 60 Saudi medicinal plants with recognised antiviral activities have been investigated and lead to the identification of nine potential inhibitors for three major druggable targets of SARS-CoV-2. The obtained candidates from the virtual screening were subjected to MD simulation to confirm their binding and dynamic stability at the target site. It has been demonstrated that chrysophanol 8-(6-galloylglucoside), luteolin 7-rutinoside and kaempferol 7-(6''-galloylglucoside) have the potential for further optimisation as an RdRp inhibitors. Moreover, 3,4,5-tri-O-galloylquinic acid, chrysophanol 8-(6-galloylglucoside) and iso-mulbel-rochromene showed potential to efficiently bind and inhibit the viral 3CL pro . Withanolide A, isocodonocarpine and calonysterone have also shown promising results in targeting the PL pro . Future in vitro and in vivo evaluations of these compounds can confirm their antiviral activity and subsequently their potential to serve as drug candidates for further designing of target-specific anti-COVID-19 agents.

The authors claim that the researchers in this study have no conflict of interest.

Moroccan Medicinal plants as inhibitors of COVID-19: Computational investigations

Traditional use of ethnomedicinal native plants in the Kingdom of Saudi Arabia

Pharmacoinformatics and molecular dynamic simulation studies to identify potential small-molecule inhibitors of WNK-SPAK/OSR1 signaling that mimic the RFQV motifs of WNK kinases

Coronavirus main proteinase (3CLpro) structure: basis for design of anti-SARS drugs

In vitro evaluation of novel antiviral activities of 60 medicinal plants extracts against hepatitis B virus. Experimental and therapeutic medicine

An open chemical toolbox

The SARS-coronavirus papain-like protease: structure, function and inhibition by designed antiviral compounds

Hydrogen Bonds in Protein-Ligand Complexes

Emerging coronaviruses: genome structure, replication, and pathogenesis

Small-molecule library screening by docking with PyRx

Pymol: An open-source molecular graphics tool

Antiviral effect of polyphenol rich plant extracts on herpes simplex virus type 1. Food Science and Human Wellness

Coronavirus (COVID-19) update: FDA issues emergency use authorization to decontaminate millions of N95 respirators

Structure of the RNA-dependent RNA polymerase from COVID-19 virus

Putative Inhibitors of SARS-CoV-2 Main Protease from A Library of Marine Natural Products: A Virtual Screening and Molecular Modeling Study

Using AutoDock 4 with AutoDocktools: a tutorial. The Scripps Research Institute

Nsp3 of coronaviruses: Structures and functions of a large multi-domain protein

Learning from the past: possible urgent prevention and treatment options for severe acute respiratory infections caused by 2019-nCoV

The potential chemical structure of anti-SARS-CoV-2 RNA-dependent RNA polymerase

Natural product-derived phytochemicals as potential agents against coronaviruses: a review. Virus Research

A taxonomically-driven approach to development of potent, broadspectrum inhibitors of coronavirus main protease including SARS-CoV-2 (COVID-19)

Bioactive compounds and health benefits of edible Rumex species-A review

Bioactive Compounds from Marrubium Vulgare L. based on in vitro Antioxidant Activity

Structures of the Middle East respiratory syndrome coronavirus 3C-like protease reveal insights into substrate specificity

The crystal structure of papain-like protease of SARS CoV-2

Tanshinones as selective and slow-binding inhibitors for SARS-CoV cysteine proteases

UCSF Chimera-a visualization system for exploratory research and analysis

Recent development of 3C and 3CL protease inhibitors for anti-coronavirus and anti-picornavirus drug discovery

A noncovalent class of papain-like protease/deubiquitinase inhibitors blocks SARS virus replication

The newly emerged SARS-like coronavirus HCoV-EMC also has an" Achilles' heel": current effective inhibitor targeting a 3C-like protease

High-throughput screening and identification of potent broad-spectrum inhibitors of coronaviruses

Mechanisms and enzymes involved in SARS coronavirus genome expression

AutoDock Vina: improving the speed and accuracy of docking with a new scoring function, efficient optimization, and multithreading

Structural basis of SARS-CoV-2 3CLpro and anti-COVID-19 drug discovery from medicinal plants

Remdesivir and chloroquine effectively inhibit the recently emerged novel coronavirus (2019-nCoV) in vitro

Evolution of the novel coronavirus from the ongoing Wuhan outbreak and modeling of its spike protein for risk of human transmission

Structural basis for inhibition of the RNA-dependent RNA polymerase from SARS-CoV-2 by remdesivir

p53 degradation by a coronavirus papain-like protease suppresses type I interferon signaling

In silico screening of Chinese herbal medicines with the potential to directly inhibit 2019 novel coronavirus

Molecular biology of severe acute respiratory syndrome coronavirus

SwissParam: a fast force field generation tool for small organic molecules

Coronaviruses-drug discovery and therapeutic options

Authors would like to thank Prince Sattam Bin Abdulaziz University for their support and providing the necessary computational tools to conduct this study.