key: cord-312160-2820aftb
authors: Ibrahim, Mahmoud A.A.; Abdelrahman, Alaa H.M.; Hussien, Taha A.; Badr, Esraa A.A.; Mohamed, Tarik A.; El−Seedi, Hesham R.; Pare, Paul W.; Efferth, Thomas; Hegazy, Mohamed Elamir F.
title: In silico Drug Discovery of Major Metabolites from Spices as SARS-CoV-2 Main Protease Inhibitors
date: 2020-10-08
journal: Comput Biol Med
DOI: 10.1016/j.compbiomed.2020.104046
sha: 
doc_id: 312160
cord_uid: 2820aftb

Coronavirus Disease 2019 (COVID-19) is an infectious illness caused by Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2), originally identified in Wuhan, China (December 2019) and has since expanded into a pandemic. Here, we investigate metabolites present in several common spices as possible inhibitors of COVID-19. Specifically, 32 compounds isolated from 14 cooking seasonings were examined as inhibitors for SARS-CoV-2 main protease (M(pro)), which is required for viral multiplication. Using a drug discovery approach to identify possible antiviral leads, in silico molecular docking studies were performed. Docking calculations revealed a high potency of salvianolic acid A and curcumin as M(pro) inhibitors with binding energies of −9.7 and −9.2 kcal/mol, respectively. Binding mode analysis demonstrated the ability of salvianolic acid A and curcumin to form nine and six hydrogen bonds, respectively with amino acids proximal to M(pro)'s active site. Stabilities and binding affinities of the two identified natural spices were calculated over 40 ns molecular dynamics simulations and compared to an antiviral protease inhibitor (lopinavir). Molecular mechanics-generalized Born surface area energy calculations revealed greater salvianolic acid A affinity for the enzyme over curcumin and lopinavir with energies of −44.8, −34.2 and −34.8 kcal/mol, respectively. Using a STRING database, protein-protein interactions were identified for salvianolic acid A included the biochemical signaling genes ACE, MAPK14 and ESR1; and for curcumin, EGFR and TNF. This study establishes salvianolic acid A as an in silico natural product inhibitor against the SARS-CoV-2 main protease and provides a promising inhibitor lead for in vitro enzyme testing.

Coronaviruses belong to the Coronaviridae family and are named for distinctive protein spikes covering the virus' outer membrane surface. Several members of the family are known to cause respiratory tract infections in humans ranging from mild common colds to severe SARS and MERS infections [1, 2] . Coronavirus Disease 2019 (COVID- 19) was first observed in Wuhan Province and identified by the Chinese Centre for Disease Control and Prevention as Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) [3, 4] . The viral genome harbors 11 genes encoding 29 proteins and peptides;

(www.ncbi.nlm.nih.gov/nuccore/NC_045512.2?report=graph). Four proteins constitute the viral structure, including the spike or S protein [5] . In SARS-CoV-2, the S protein binds to an angiotensin-converting enzyme 2 (ACE2), a necessary step for viral entry into the host cell.

Studies thus far indicate that the virus' S protein binds stronger to ACE2 than the one of SARS-CoV, providing a rationale why COVID-19 so easily spreads and is highly infectious.

Another group of SARS-CoV-2 proteins controls how the virus replicates as well as avoids the host's immune system. These non-structural proteins initially expressed as two large polyproteins are processed into 16 peptide components. The main protease (M pro or 3CLpro), cleaves the polyproteins into 11 fragments, whose structures were recently elucidated and an inhibitor that blocks the M pro catalytic activity identified [6] . From this work, M pro appears to be a promising target for designing small molecule inhibitors.

COVID-19 rapidly spreads due to the global mobility of humans and is currently present in more than 200 countries. Patients mainly suffer from fever, dry cough, labored breathing, and bilateral lung infiltrates. The causative agent is diagnosed from throat or nasal swabs J o u r n a l P r e -p r o o f with nucleic acid sequence similarity. Rapid disease spreading coupled with high mortality rates makes COVID-19 a major global public health threat [7] . Recently, emergency use of remdesivir has been issued by the U.S. Food and Drug Administration for treatment of COVID-19. With few treatment options available, there is an urgent need to seek out effective strategies for prophylaxis for such viral outbreaks. Using experimental methods of drug discovery is time-consuming and costly. Therefore, structure-based computational modeling of ligand-receptor interactions can be used to identify potential M pro inhibitors to block viral replication.

Herbal extracts and spices are natural immune boosters and/or anti-infective agents currently utilized in many parts of the world [8] . In traditional folk medicine, spices, botanical detoxifiers [9] , antioxidants [10] and plant haematinics [11] are used as antiviral mediators to prevent/minimize disease. Low toxicity makes such metabolites well suited as drug leads for viral diseases such as COVID-19. In this study, selected spices with documented, biologically activity (e.g. cinnamon, clove, ginger, mustard and others) were exemplarily chosen to generate a metabolite library for the screening of M pro -specific drug candidates with presumable effectiveness against COVID-19.

The resolved crystal structure of the main protease (M pro ) of SARS-CoV-2 in complex with N3 inhibitor (PDB code: 6LU7 [12] ) was used for molecular docking as well as molecular dynamics calculations. Water and spectator ions were deleted. H++ server was J o u r n a l P r e -p r o o f used to study the protonation state of M pro and to add all missing hydrogen atoms [13] . In H++ calculations, the following physical conditions were applied: pH=6.5, internal dielectric=10, external dielectric= 80 and salinity=0.15.

The chemical structures of the 32 investigated natural spices were retrieved from the PubChem database and their 3D structures were generated using Omega2 software [14, 15] .

All generated structures were minimized using Merck Molecular Force Field 94 (MMFF94S) with the assistance of available software (SZYBKI) [16] . The 2D chemical structures of the investigated compounds are illustrated in Table 1 .

For molecular docking calculations, AutoDock4.2.6 software was utilized [17] . The pdbqt file of SARS-CoV-2 M pro was prepared according to the AutoDock protocol [18] . In AutoDock4.2.6, default parameters were employed, except the numbers of genetic algorithm (GA) run and energy evaluations (eval). GA and eval were set to 250 and 25,000,000, respectively. The grid was defined to cover the active site of the SARS-CoV-2 M pro . The grid size and spacing value were 60 Å × 60 Å × 60 Å and 0.375 Å, respectively. The grid center coordinates were −13.069, 9.740, 68.490 (XYZ assignments, respectively). The atomic charges of studied natural spices were assigned using the Gasteiger method [19] . The predicted binding poses for each compound were processed by the built-in clustering analysis 

AMBER16 software was utilized to conduct molecular dynamics (MD) simulation for the natural spices in complex with SARS-CoV-2 M pro [20] . The details of the employed MD simulations are described in Ref. [21, 22] . In brief, general AMBER force field (GAFF) [23] and AMBER force field 14SB [24] were applied to describe spices compounds and M pro , respectively. Restrained electrostatic potential (RESP) approach [25] was utilized to assign the atomic partial charges of the natural spices using Gaussian09 software [26] . Docked spice-M pro complexes were water solvated with 15 Å distances between the box edge and atoms of the spice-M pro complexes. Solvated spice-M pro complexes were minimized by 5000 steps and afterward smoothly heated from 0 K to 300 K over a brief interval (50 ps). Using periodic boundary conditions and NPT ensemble, the spice-M pro systems were simulated for 10 ns of equilibration and 40 ns of production. All molecular dynamics simulations were carried out with pmemd.cuda implemented in AMBER16. All molecular docking and molecular dynamics calculations were performed on CompChem GPU/CPU cluster (hpc.compchem.net).

The binding energies of the investigated spices compounds with SARS-CoV-2 M pro were estimated using molecular mechanical-generalized Born surface area (MM-GBSA) approach with modified GB model (igb=2) implemented in AMBER16 software [27] . For the MM-GBSA calculations, uncorrelated snapshots were collected over the production run, and a single-trajectory approach was employed, in which compound, receptor, and complex J o u r n a l P r e -p r o o f coordinates were retrieved from a single trajectory. The binding energy (ΔG binding ) was estimated as follows:

where the energy term (G) is estimated as: 

The physicochemical parameters of the most promising natural spices as SARS-CoV-2 M pro inhibitors were predicted using the online Molinspiration cheminformatics software %ABS was estimated as follows [28] :

J o u r n a l P r e -p r o o f

The online web-based tools of SwissTargetPredicition (http://www.swisstargetprediction.ch) were applied to predict the biological targets for the most promising natural spices as SARS-CoV-2 M pro inhibitors. The DisGeNET online database (https://www.disgenet.org) was utilized to collect the available database for SARS diseases. Venn Diagram was designed using InteractiVenn online tool [29] . Protein-protein interaction (PPI) network was generated using a STRING functional database for top predicted targets [30] . Cytoscape 3.8.0 was employed to investigate target-function relation based on the network topology [31] .

Lack of treatments against COVID-19 pinpoints a critical need to systematically screen and identify compounds that can block viral reproduction. Since the main protease of SARS-CoV-2 (M pro ) plays a critical role in the viral replication process, structure-based computational modeling of ligand-receptor interactions and molecular dynamics has been used to screen metabolites from common spices as potential M pro inhibitors. Indeed several herbal plants have already been reported as antiviral entities against hepatitis B, respiratory syncytial virus and influenza [32] . (Table 1) . Most docked natural products shared the same binding modes, forming hydrogen bonds with key amino acid residues in the active site such as THR190, GLY143, CYS145, and GLU166. 2D binding modes for each of the compounds are displayed in Figure S1 . The peptidomimetic molecule lopinavir, which functions as an antiretroviral protease inhibitor against HIV was used as a positive control [33, 34] as it has recently been clinically Figure S1 . b No hydrogen bond was observed.

Since the reliability of ligand-enzyme binding energies using molecular docking scores have been questioned due to complicating environmental factors such as a lack of ligand-receptor flexibility, solvent effects, and dynamics [37, 38] Table 2 . For salvianolic acid A, binding energy was dominated by E ele interactions with an average value of −65.5 kcal/mol which was three times higher than that of lopinavir and curcumin, with an average value of −26.1 and −19.8 kcal/mol, respectively. This is attributed to a higher number of hydrogen bonds for salvianolic acid A with the key amino acids inside M pro 's active site, compared to lopinavir or curcumin (Table 1) 

While 

Drug-likeness is a qualitative measure utilized in drug discovery to evaluate pharmacokinetic properties such as oral bioavailability. Physicochemical parameters were evaluated using Molinspiration cheminformatics, (http://www.molinspiration.com) online software calculation toolkit. The predicted parameters are summarized in 

Salvianolic acid A and curcumin protein targets were predicted and classified using a SwissTargetPrediction ( Figure S2 ). One hundred and seventeen genes were identified using DisGeNET online tools for Severe Acute Respiratory Syndrome diseases (SARS, C1175175). Utilizing Venn diagram comparison analysis, commonly shared genes for salvianolic acid A included ACE, CASP3, CASP1, ESR1 and MAPK14, and for curcumin TNF, EGFR and ADAM17 ( Figure 5 ). Angiotensin-converting enzyme 2 (ACE2) is a host protein and the receptor for SARS-CoV-2 entry [39] . MAPK14 inhibition is predicted to block the ACE2 signaling pathway, and in turn, reduce cell internalization of SARS-CoV-2.

For SARS-S ADAM17-dependent shedding of ACE2, a process coupled with TNF-α production, reduced viral reproduction [40] . Salvianolic acid A and curcumin predicted J o u r n a l P r e -p r o o f genes targets were also analyzed via a STRING PPI network and visualized by Cytoscape 3.8.0. The top 10 scored genes for salvianolic acid A included ACE, MAPK14 and ESR1 and for curcumin EGFR and TNF (Table S1 ). 

The COVID-19 pandemic has had a catastrophic impact on human health and global economies. SARS-CoV-2 main protease (M pro ) may well prove to be the Achilles heel of

Coronavirus genomics and bioinformatics analysis

MERS, SARS and other coronaviruses as causes of pneumonia

A Novel Coronavirus from Patients with Pneumonia in China

World Health Organization. WHO director-general's remarks at the media briefing on 2019-nCoV on 11

Composition and divergence of coronavirus spike proteins and host ACE2 receptors predict potential intermediate hosts of SARS-CoV-2

Crystal structure of SARS-CoV-2 main protease provides a basis for design of improved alpha-ketoamide inhibitors

Live): 2,418,429 Cases and 165,739 Deaths from COVID−19 Virus Pandemic-Worldometer

Stay Safe: Helpful Herbal remedies in COVID-19 infection

Hidden Treasures of Ethnobotanical Medicine. A Faculty Lecture delivered at the Faculty of Science

In vitro anti-radical activities of extracts of Solanum nigrum (L.) from South Africa

Nutritional composition of ten ethnobotanicals used for the treatment of anaemia in Southwest Nigeria

Structure of M(pro) from SARS-CoV-2 and discovery of its inhibitors

H++: a server for estimating pKas and adding missing hydrogens to macromolecules

Conformer Generation with OMEGA: Algorithm and Validation Using High Quality Structures from the

Protein Databank and Cambridge Structural Database

AutoDock4 and AutoDockTools4: Automated docking with selective receptor flexibility

Computational protein-ligand docking and virtual drug screening with the AutoDock suite

Iterative Partial Equalization of Orbital Electronegativity -a Rapid Access to Atomic Charges

In-silico drug repurposing and molecular dynamics puzzled out potential SARS-CoV-2 main protease inhibitors

Natural-like products as potential SARS-CoV-2 M(pro) inhibitors: in-silico drug discovery

Development and testing of a general amber force field

Simmerling, ff14SB: Improving the Accuracy of Protein Side Chain and Backbone Parameters from ff99SB

A well-behaved electrostatic potential based method using charge restraints for deriving atomic charges: the RESP model

Gaussian 09

Combined molecular mechanical and continuum solvent approach (MM-PBSA/GBSA) to predict ligand binding

Rate-limited steps of human oral absorption and QSAR studies

InteractiVenn: a web-based tool for the analysis of sets through Venn diagrams

Anti-colorectal cancer targets of resveratrol and biological molecular mechanism: Analyses of network pharmacology, human and experimental data

Cytoscape: a software environment for integrated models of biomolecular interaction networks

Antiviral natural products and herbal medicines

Lopinavir/ritonavir in the treatment of HIV-1 infection: a review

The metabolic effects of lopinavir/ritonavir in HIV-negative men

A Trial of Lopinavir-Ritonavir in Adults Hospitalized with Severe Covid-19

Clinical efficacy of lopinavir/ritonavir in the treatment of Coronavirus disease 2019

Binding Affinity via Docking: Fact and Fiction

Software for molecular docking: a review

SARS-CoV-2 Cell Entry Depends on ACE2 and TMPRSS2 and Is Blocked by a Clinically Proven Protease Inhibitor

ADAM17 inhibition may exert a protective effect on COVID-19