In Silico Mining of Terpenes from Red-Sea Invertebrates for SARS-Cov-2 Main Protease (Mpro) Inhibitors

Article In Silico Mining of Terpenes from Red-Sea Invertebrates for SARS-CoV-2 Main Protease (Mpro) Inhibitors

Mahmoud A. A. Ibrahim 1,* , Alaa H. M. Abdelrahman 1 , Tarik A. Mohamed 2 , Mohamed A. M. Atia 3, Montaser A. M. Al-Hammady 4 , Khlood A. A. Abdeljawaad 1 , Eman M. Elkady 4, Mahmoud F. Moustafa 5,6, Faris Alrumaihi 7 , Khaled S. Allemailem 7 , Hesham R. El-Seedi 8,9,10,*, Paul W. Paré 11 , Thomas Efferth 12 and Mohamed-Elamir F. Hegazy 2,12,*

1 Computational Chemistry Laboratory, Chemistry Department, Faculty of Science, Minia University, Minia 61519, Egypt; [email protected] (A.H.M.A.); [email protected] (K.A.A.A.) 2 Chemistry of Medicinal Plants Department, National Research Centre, 33 El-Bohouth St., Dokki, Giza 12622, Egypt; [email protected] 3 Molecular Genetics and Genome Mapping Laboratory, Genome Mapping Department, Agricultural Genetic Engineering Research Institute (AGERI), Agricultural Research Center (ARC), Giza 12619, Egypt; [email protected] 4 National Institute of Oceanography & Fisheries, NIOF, Cairo 11516, Egypt; [email protected] (M.A.M.A.-H.); [email protected] (E.M.E.) 5 Department of Biology, College of Science, King Khalid University, Abha 9004, Saudi Arabia; [email protected] 6 Department of Botany & Microbiology, Faculty of Science, South Valley University, Qena 83523, Egypt Citation: Ibrahim, M.A.A.; 7 Department of Medical Laboratories, College of Applied Medical Sciences, Qassim University, Abdelrahman, A.H.M.; Mohamed, Buraydah 51452, Saudi Arabia; [email protected] (F.A.); [email protected] (K.S.A.) 8 T.A.; Atia, M.A.M.; Al-Hammady, Department of Molecular Biosciences, The Wenner-Gren Institute, Stockholm University, S-106 91 Stockholm, Sweden M.A.M.; Abdeljawaad, K.A.A.; 9 Department of Chemistry, Faculty of Science, El-Menoufia University, Shebin El-Kom 32512, Egypt Elkady, E.M.; Moustafa, M.F.; 10 International Research Center for Food Nutrition and Safety, Jiangsu University, Zhenjiang 212013, China Alrumaihi, F.; Allemailem, K.S.; et al. 11 Department of Chemistry & Biochemistry, Texas Tech University, Lubbock, TX 79409, USA; [email protected] In Silico Mining of Terpenes from 12 Department of Pharmaceutical Biology, Institute of Pharmaceutical and Biomedical Sciences, Red-Sea Invertebrates for Johannes Gutenberg University, Staudinger Weg 5, 55128 Mainz, Germany; [email protected] SARS-CoV-2 Main Protease (Mpro) * Correspondence: [email protected] (M.A.A.I.); [email protected] (H.R.E.-S.); Inhibitors. Molecules 2021, 26, 2082. [email protected] (M.-E.F.H.); Tel.: +2-010-241-61-444 (M.A.A.I.); +46-73-566-8234 (H.R.E.-S.); https://doi.org/10.3390/ +2-033-371-635 (M.-E.F.H.) molecules26072082 Abstract: Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is the causative agent for Academic Editor: Pedro Silva the COVID-19 pandemic, which generated more than 1.82 million deaths in 2020 alone, in addition to 83.8 million infections. Currently, there is no antiviral medication to treat COVID-19. In the Received: 4 March 2021 search for drug leads, marine-derived metabolites are reported here as prospective SARS-CoV-2 Accepted: 30 March 2021 inhibitors. Two hundred and twenty-seven terpene natural products isolated from the biodiverse Published: 5 April 2021 Red-Sea ecosystem were screened for inhibitor activity against the SARS-CoV-2 main protease (Mpro) using molecular docking and molecular dynamics (MD) simulations combined with molecular Publisher’s Note: MDPI stays neutral mechanics/generalized Born surface area binding energy calculations. On the basis of in silico with regard to jurisdictional claims in analyses, six terpenes demonstrated high potency as Mpro inhibitors with ∆G ≤ −40.0 kcal/mol. published maps and institutional affil- binding iations. The stability and binding affinity of the most potent metabolite, erylosides B, were compared to the human immunodeficiency virus protease inhibitor, lopinavir. Erylosides B showed greater binding pro affinity towards SARS-CoV-2 M than lopinavir over 100 ns with ∆Gbinding values of −51.9 vs. −33.6 kcal/mol, respectively. Protein–protein interactions indicate that erylosides B biochemical signaling shares gene components that mediate severe acute respiratory syndrome diseases, including Copyright: © 2021 by the authors. Licensee MDPI, Basel, Switzerland. the cytokine- and immune-signaling components BCL2L1, IL2, and PRKC. Pathway enrichment This article is an open access article analysis and Boolean network modeling were performed towards a deep dissection and mining of distributed under the terms and the erylosides B target–function interactions. The current study identifies erylosides B as a promising conditions of the Creative Commons anti-COVID-19 drug lead that warrants further in vitro and in vivo testing. Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

Molecules 2021, 26, 2082. https://doi.org/10.3390/molecules26072082 https://www.mdpi.com/journal/molecules Molecules 2021, 26, 2082 2 of 22

Keywords: drug discovery; marine natural products; molecular docking; molecular dynamics; SARS-CoV-2 main protease; virtual drug screening

1. Introduction Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) belongs to the family Coronaviridae (genus Betacoronavirus; subgenus Sarbecovirus) with a viral envelope encapsulating a single-stranded RNA genome. While most coronaviruses infect mammals and birds, the animal host responsible for the transmission of SARS-CoV-2 to men has yet to be identified [1–3]. The SARS-CoV-2 genome contains 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 responsible for binding with the host cell receptor [4]. The S protein binds to the angiotensin-converting enzyme 2 (ACE2), which is a necessary step for viral entry into the cell. Two currently employed anti-SARS- CoV-2 vaccines utilize nanoparticle-encapsulated mRNA coding for the spike protein. These vaccines activate human immune responses to target the S protein, and about 95% of vaccinated people are protected from severe courses of the Coronavirus disease 2019 (COVID-19) [5]. In the case of more contagious variants such as a strain originating from Brazil (known as P.1), the genome has 17 unique mutations, including three in the receptor-binding domain of the spike protein. Mutations in the S protein are thought to be responsible for the variable ability of the spike protein to bind to ACE2, and specifically, the P.1 mutant reveals increased affinity and penetration into the host. Recently, an increasing number of escape mutations in the spike protein threaten the effectiveness of vaccines [6]. In seeking alternative approaches to vaccination, other SARS-CoV-2 proteins can be investigated that control the virus replication. Essential nonstructural proteins are initially expressed as two large polyproteins that are then processed into 16 peptide components. The main protease (Mpro) initially cleaves the polyproteins into 11 fragments. Inhibitors that block the catalytic Mpro activity effectively interrupt the viral replication. One such inhibitor has been recently identified [7]. The crystal structure of SARS-CoV-2 main protease provides a basis for the design of improved alpha-ketoamide inhibitors [7] and Mpro appears to be a promising target for designing novel small molecule inhibitors. Since that original study, several promising inhibitor leads have been identified using in silico enzyme modeling [8]. Structure-based computational modeling of ligand–receptor interactions was used by Ibrahim et al. to identify potential Mpro inhibitors [9–13]. Natural products hold a vital role in discovering novel and effective therapeutics to combat the present COVID-19 pandemic. Among natural products, flavonoids, alkaloids, and terpenoids have attracted great attention as prospective SARS-CoV-2 inhibitors [14–16]. Recogniz- ing that marine invertebrates are promising organisms for biologically active metabolites including anti-inflammatory, antibacterial, antifungal, antimalarial, antitumor, and antiviral activity [17,18], here biologically active terpene metabolites identified from a coral reef community unique to the Red Sea [19] were screened for in silico binding affinities against SARS-CoV-2 Mpro. Previously characterized metabolites from this natural-product pool include alismol and aromadendrane sesquiterpenes derived from Litophyton arbore- tum [20] that exhibit inhibitory activity against the HIV-1 protease (HIV-1 PR) (IC50 7 µM); palustrol, a sesquiterpene from Sarcophyton trocheliophorum that has antibacterial activity (MIC 6.6–11.1 µM) [21]; and 12(S)-Hydroperoxylsarcoph-10-ene, a cembrane diterpene from Sarcophyton glaucum that was reported to exhibit potent anticancer activity via the inhibition of Cyp1A activity (p < 0.01) with IC50 values of 2.7 nM [22]. On the basis of the predicted docking scores, the most potent inhibitors are submitted to molecular dynamics (MD) simulations combined with binding energy calculations using a molecular mechanics/generalized Born surface area approach. Molecules 2021, 26, 2082 3 of 22

2. Results and Discussion Since the main protease (Mpro) of SARS-CoV-2 plays an indispensable role in viral reproduction, small molecules were screened based on in silico molecular docking calculations and MD simulations for prospective Mpro inhibitors. Marine natural products identiﬁed from the Red Sea provided the source for metabolite screening.

2.1. Molecular Docking Two hundred and twenty-seven terpene natural products isolated from the biodiverse Red-Sea ecosystem were screened against the SARS-CoV-2 main protease (Mpro) using molecular docking technique. Molecular docking calculations resulted in 27 of the screened compounds exhibiting a higher binding affinity than lopinavir: an inhibitor of SARS-CoV-2 main protease (Mpro) that was proposed as a treatment for COVID-19 on the basis of in vitro activity, preclinical studies, and observational studies [23]. While docking scores ranged from −4.3 to −12.3 kcal/mol, 12% of the compounds scored below −9.8 kcal/mol (Table S1). AutoDock4.2.6 software was utilized to carry out all molecular docking calculations. Binding affinities, 2D chemical structures, and features of the 27 most promising natural products towards SARS-CoV-2 Mpro are summarized in Table1. 2D docking positions with proximal amino acid residues within the Mpro active site are depicted in Figure S1. Most of these compounds demonstrate similar Mpro binding modes within the binding pocket, forming hydrogen bonds with CYS145, HIS164, and GLU166, which can account for the high binding affinities (Table1 and Figure S1). The 2D and 3D representations of the interactions of the top three potent marine natural products (MNPs) and lopinavir with key amino acid residues of SARS-CoV-2 Mpro are depicted in Figure1 and Figure S2, respectively. Depresosterol (190) isolated from Lobophytum depressum, exhibited the highest binding affinity of the compounds screened against SARS-CoV-2 Mpro, with a docking score of −12.3 kcal/mol. in silico Mpro binding in the active site indicated that the methanolic hydroxyl group exhibited two hydrogen bonds with a backbone carboxylate of GLU166 with bond lengths of 1.99 and 2.55 Å, respectively (Figure1 and Table1). In addition, the hydroxyl unit of 2-methylpropan-2-ol affords three hydrogen bonds with a backbone NH and carbonyl group of ASN142 with bond lengths of 2.24, 2.68, and 2.04 Å, respectively (Figure1 and Table1). Moreover, the hydroxy group of 2-propanol exhibited a hydrogen bond with the backbone carbonyl group of ASN142 with a bond length of 1.96 Å (Figure1 , Figure S2 and Table1). The oxygen of the oxirane ring interacted with the backbone imidazole ring of HIS41, and the thiol group of CYS145 with bond lengths of 2.17 and 2.70 Å, respectively (Figure1 and Table1). The hydroxy group of the cyclohexanol ring contributed two hydrogen bonds with NH and the carbonyl group of TYR26 with bond lengths of 2.15 and 2.66 Å, respectively (Figure1 and Table1). 3β-25-Dihydroxy-4-methyl-5α,8α-epidioxy-2-ketoergost-9-ene (178) isolated from Sinularia candidula, exhibited the second highest binding affinity of the compounds screened against SARS-CoV-2 Mpro with a docking score of −12.2 kcal/mol. in silico Mpro binding in the active site indicated that the hydroxy group of the hydroxycyclohexanone ring participates in four hydrogen bonds with the backbone carbonyl of LEU141, OH and NH of SER144, and NH of CYS145 with bond lengths of 2.08, 1.97, 2.28, and 2.49 Å, respectively (Figure1 and Table1). Moreover, the carbonyl group of the hydroxycyclohexanone ring demonstrates two hydrogen bonds with an imidazole ring of HIS163 and backbone OH of SER144 with bond lengths of 2.08 and 2.66 Å, respectively (Figure1 and Table1). The hydroxy group of 2-methylpropan-2-ol displays two hydrogen bonds with a backbone carbonyl group of THR190 and NH of GLN192 with bond lengths of 1.80 and 2.26 Å, respectively (Figure1 and Table1). Molecules 2021, 26, x FOR PEER REVIEW 4 of 22 Molecules 2021, 26, x FOR PEER REVIEW 4 of 22

Molecules 2021, 26, x FOR PEER REVIEWhydrogen bonds with a backbone OH of TYR54, imidazole ring of HIS41, SH4 ofof CYS145,22 hydrogen bonds with a backbone OH of TYR54, imidazole ring of HIS41, SH of CYS145, imidazole ring of HIS163, NH of ASN142, NH of GLY143, and NH2 of GLN189 with bond imidazole ring of HIS163, NH of ASN142, NH of GLY143, and NH2 of GLN189 with bond lengths of 3.05, 1.96, 2.04, 1.93, 1.75, 2.82, and 1.91 Å, respectively (Figure 1 and Table 1). lengths of 3.05, 1.96, 2.04, 1.93, 1.75, 2.82, and 1.91 Å, respectively (Figure 1 and Table 1). Moreover, the oxygen of the (cyclohexyloxy) cyclohexane ring forms a hydrogen bond Moreover,hydrogen the bonds oxygen with aof backbone the (cyclohexyloxy) OH of TYR54, cyclohexaneimidazole ring ring of HIS41, forms SH a ofhydrogen CYS145, bond with the backbone NH of GLU166 with a bond length of 2.37 Å (Figure 1 and Table 1). withimidazole the backbone ring of HIS163, NH of NH GLU166 of ASN142, with NHa bond of GLY143, length and of NH2.372 ofÅ GLN189(Figure with1 and bond Table 1). Thelengths oxygen of 3.05,atom 1.96, of the2.04, methoxy 1.93, 1.75, group2.82, an isd 1.91also Å,hydrogen respectively bound (Figure to 1the and NH Table of 1).ASN142 The oxygen atom of the methoxy group is also hydrogen bound to the NH of ASN142 with a bond length of 2.96 Å (Figure 1 and Table 1). Compared with 190, 178, and 226, withMoreover, a bond thelength oxygen of 2.96of the Å (cyclohexyloxy)(Figure 1 and cyclohexaneTable 1). Compared ring forms with a hydrogen 190, 178, bond and 226, lopinavirwith the presented backbone NHa similar of GLU166 binding with affinity a bond lengthtowards of 2.37Mpro Å with (Figure a docking 1 and Table score 1). of −9.8 lopinavir presented a similar binding affinity towards Mpro with a docking score of −9.8 kcal/mol.The oxygen Scrutinizing atom of the the methoxy binding groupmode isof also lopinavir hydrogen inside bound the toactive the NH site ofof ASN142Mpro revealed kcal/mol. Scrutinizing the binding mode of lopinavir inside the active site of Mpro revealed thatwith the a NH bond of lengththe tetrahydro-1-methylpyrimidin-2(1H)-one of 2.96 Å (Figure 1 and Table 1). Compared ring with exhibited 190, 178, twoand 226,hydrogen that the NH of the tetrahydro-1-methylpyrimidin-2(1H)-onepro ring exhibited two hydrogen bondslopinavir with presenteda backbone a similar hydroxy binding group affinity of SER144 towards and M a withcarbonyl a docking group score of LEU141 of −9.8 with bondskcal/mol. with Scrutinizing a backbone the hydroxy binding groupmode of of lopinavir SER144 inside and a the carbonyl active site group of M proof revealed LEU141 with bond lengths of 3.09 and 1.96 Å, respectively. bondthat lengths the NH of the3.09 tetrahydro-1-methylpyrimidin-2(1H)-one and 1.96 Å, respectively. ring exhibited two hydrogen Molecules 2021, 26, 2082 4 of 22 bonds with a backbone hydroxy group of SER144 and a carbonyl group of LEU141 with Table 1. Estimated docking scores, 2D chemical structures, and binding features for lopinavir and the top 27 potent marine Table 1. Estimated docking scores,bond 2D lengths chemical of 3.09 structures, and 1.96 and Å, bindrespectively.ing features for lopinavir and the top 27 potent marine natural products (MNPs) towards SARS-CoV-2 main protease (Mpro). natural products (MNPs) towards SARS-CoV-2 main protease (Mpro). Table 1. Estimated docking scores, 2D chemical structures, and binding features for lopinavir and the top 27 potent marine Table 1. Estimated docking scores, 2D chemical structures, and binding features for lopinavir and the top 27 potent marine natural products (MNPs) towards SARS-CoV-2 main protease (Mpro). Docking pro Docking naturalMNP products Name (MNPs) Plant towards Source SARS-CoV-2 2D mainChemical protease Structure (M ). Score~~~ Binding Features a MNP Name Plant Source 2D Chemical Structure DockingScore~~~ Binding Features a (kcal/mol)Docking MNP Name Plant Plant Source 2D Chemical Structure Score~~~(kcal/mol) Binding Features a MNP Name 2D Chemical Structure Score Binding Features a Source (kcal/mol)(kcal/mol)

HIS164 (2.62 Å),~~~ HIS164 (2.62 Å),~~~ HIS164 (2.62 Å),~~~ GLY143HIS164 (2.01 (2.62Å),~~~ Å), Lopinavir --- b −9.8 GLY143 (2.01 Å),~~~ Lopinavir --- b −9.8 GLY143 (2.01GLY143 Å),~~~ (2.01 Å), LopinavirLopinavir — b --- b −9.8 −9.8 LEU141 (1.96 Å),~~~ LEU141LEU141 (1.96LEU141 (1.96Å),~~~ Å),~~~ (1.96 Å), SER144SER144 (3.09 (3.09 Å) Å) SER144SER144 (3.09 (3.09Å) Å)

THR26 (2.15, 2.66 Å),~~~ THR26THR2 (2.15,6 (2.15, 2.66 Å),~~~2.66 Å),~~~ HIS41HIS41 (2.17THR26 (2.17 Å),~~~ (2.15, Å),~~~ 2.66 Å), HIS41 (2.17 Å),~~~ Depresosterol~~~Depresosterol~~~ HIS41 (2.17 Å), DepresosterolDepresosterol~~~ L. depressuL. depressumm −12.3−12 .3 CYS145CYS145 (2.70 (2.70Å),~~~ Å),~~~ (190) L. depressumL. depressum −12.3 −12.3 CYS145CYS145 (2.70 Å),~~~ (2.70 Å), (190)(190) (190) ASN142ASN142 (1.96, (1.96, 2.04, 2.04,2.24 Å),~~~ 2.24 Å),~~~ ASN142ASN142 (1.96, 2.04, (1.96, 2.24 2.04, Å),~~~ 2.24 Å), GLU166GLU166 (1.99,GLU166 (1.99, 2.55 (1.99, Å) 2.55 2.55 Å) Å) LEU141GLU166 (2.08 (1.99, Å),~~~ 2.55 Å) LEU141 (2.08 Å),~~~ 3β-25-Dihydroxy-4-me- SER144 (1.97,LEU1 2.28,41 (2.08 2.66 Å),~~~Å),~~~ LEU141 (2.08 Å), 3β-25-Dihydroxy-4--25-Dihydroxy-4-me- SER144 (1.97, 2.28, 2.66 Å),~~~ 3β-25-Dihydroxy-4-me-thyl-5α,8α-epidioxy-2- Sinularia can- SER144HIS163SER144 (1.97, (2.08 Å),~~~2.28, (1.97, 2.66 2.28, Å),~~~ 2.66 Å), Moleculesmethyl-5 2021α, ,826α, -x FOR PEER REVIEW −12.2 5 of 22 thyl-5α,8α-epidioxy-2-SinulariaSinularia can- HIS163HIS163 (2.08 Å),~~~ (2.08 Å), thyl-5epidioxy-2-αketoergost-9-ene~~~,8α-epidioxy-2- Sinularia didula can- −12.2 −12.2 CYS145HIS163 (2.49 (2.08 Å),~~~ Å),~~~ Molecules 2021, 26, x FOR PEERcandidula REVIEW −12.2 CYS145 (2.49 Å), 5 of 22 ketoergost-9-eneketoergost-9-ene~~~(178) didula GLN192CYS145 (2.26 (2.49 Å),~~~ Å),~~~ ketoergost-9-ene~~~ didula CYS145GLN192 (2.49 Å),~~~ (2.26 Å), (178)(178) GLN192 (2.26 Å),~~~ (178) THR190GLN192 (1.80THR190 (2.26 Å) Å),~~~ (1.80 Å) THR190 (1.80 Å) THR190 (1.80 Å) HIS41 (1.96 Å),~~~ TYR54HIS41 (1.96(3.05 Å),~~~Å),~~~ AS142TYR54 (1.75,HIS41 (3.05 2.96 (1.96Å),~~~ Å),~~~ Å), TYR54 (3.05 Å), Erylosides B~~~ AS142GLY143 (1.75, (2.82 2.96 Å),~~~ Å),~~~ E. lendenfeldi −12.1 AS142 (1.75, 2.96 Å), ErylosidesErylosides(226) B B~~~ GLY143CYS145GLY143 (2.04(2.82 Å),~~~ (2.82 Å), E. lendenfeldiE. lendenfeldi −12.1 −12.1 (226) (226) CYS145HIS163CYS145 (1.93(2.04 Å),~~~ (2.04 Å), HIS163 (1.93 Å), GLU166HIS163GLU166 (1.93 (2.37 Å),~~~ Å),~~~ (2.37 Å), GLU166GLN189GLN189 (2.37 (1.91 Å),~~~ (1.91 Å) Å) GLN189 (1.91 Å)

GLY143 (1.74 Å),~~~ GLY143CYS145GLY143 (2.88(1.74 Å),~~~ (1.74 Å), CYS145HIS163CYS145 (2.56(2.88 Å),~~~ (2.88 Å), HIS163 (2.56 Å), SipholenolSipholenol H H~~~ S. siphonellaS. siphonella −12.0 −12.0 HIS163HIS164HIS164 (2.56(2.85 Å),~~~ (2.85 Å), Sipholenol(157)(157) H~~~ S. siphonella −12.0 MET165HIS164MET165 (2.85 (2.74 Å),~~~ Å),~~~ (2.74 Å), (157) THR190 (1.74, 2.17 Å), MET165 (2.74 Å),~~~ THR190GLN192 (1.74, 2.17 (2.11 Å), Å) THR190GLN192 (1.74, (2.11 2.17 Å) Å), CYS145GLN192 (2.34 (2.11 Å),~~~ Å) CYS145HIS41 (2.43 (2.34 Å),~~~ Å),~~~ Dahabinone A~~~ S. siphonella −11.9 GLU166HIS41 (2.01, (2.43 2.23, Å),~~~ 2.38 Å),~~~ Dahabinone(162) A~~~ S. siphonella −11.9 GLU166GLY143 (2.01, (2.12 2.23, Å),~~~ 2.38 Å),~~~ (162) GLY143 (2.12 Å),~~~ GLN189 (2.03 Å) GLN189 (2.03 Å) GLY143 (2.04 Å),~~~ GLY143CYS145 (2.89(2.04 Å),~~~ Sipholenol I~~~ S. siphonella −11.8 CYS145HIS163 (2.89 Å),~~~ Sipholenol(174) I~~~ S. siphonella −11.8 HIS163HIS164 (2.89(2.01 Å),~~~ (174) GLU166HIS164 (1.62, (2.01 2.03, Å),~~~ 2.52 Å)

GLU166 (1.62, 2.03, 2.52 Å)

TYR54 (2.24, 2.55 Å),~~~ Lobophytosterol~~~ ASN142TYR54 (2.24, (1.91, 2.55 2.75 Å),~~~ Å),~~~ L. depressum −11.5 Lobophytosterol~~~(188) ASN142GLU166 (1.91,(1.86, 2.752.66 Å),~~~ L. depressum −11.5 (188) GLU166ASP187 (1.86, (2.94 2.66 Å) Å),~~~ ASP187 (2.94 Å) THR26 (2.69 Å),~~~ (22R,24E,28E)-5β,6β- THR26HIS41 (2.14(2.69 Å),~~~Å),~~~ Epoxy-22,28-oxido-24-(22R,24E,28E)-5β,6β- CYS145HIS41 (2.14 (2.45 Å),~~~ Å),~~~ Epoxy-22,28-oxido-24-methyl-5αcholestan- L. depressum −11.4 ARG188CYS145 (2.45(2.03 Å),~~~Å),~~~ methyl-53β,25,28-triol~~~αcholestan- L. depressum −11.4 THR190ARG188 (2.06, (2.03 2.55 Å),~~~ Å),~~~ 3β,25,28-triol~~~(191) THR190GLN192 (2.06, (2.26 2.55 Å)Å),~~~ (191) GLN192 (2.26 Å)

MoleculesMolecules 2021 2021, 26,, 26, x,, xFORx FORFOR PEER PEERPEER REVIEW REVIEWREVIEW 5 5of of 22 22 Molecules 2021, 26, x FOR PEER REVIEW 5 of 22

HIS41HIS41 (1.96 (1.96 Å),~~~ Å),~~~ HIS41 (1.96 Å),~~~ TYR54TYR54 (3.05 (3.05 Å),~~~ Å),~~~ TYR54 (3.05 Å),~~~ AS142AS142 (1.75, (1.75, 2.96 2.96 Å),~~~ Å),~~~ AS142 (1.75, 2.96 Å),~~~ ErylosidesErylosides B~~~ B~~~ GLY143GLY143 (2.82 (2.82 Å),~~~ Å),~~~ Erylosides B~~~ E.E. lendenfeldi lendenfeldi −12.1−12.1 GLY143 (2.82 Å),~~~ (226)(226) E. lendenfeldi −12.1 CYS145CYS145 (2.04 (2.04 Å),~~~ Å),~~~ (226) CYS145 (2.04 Å),~~~ HIS163HIS163 (1.93 (1.93 Å),~~~ Å),~~~ HIS163 (1.93 Å),~~~ GLU166GLU166 (2.37 (2.37 Å),~~~ Å),~~~ GLU166 (2.37 Å),~~~ GLN189GLN189 (1.91 (1.91 Å) Å) GLN189 (1.91 Å)

GLY143GLY143 (1.74 (1.74 Å),~~~ Å),~~~ Molecules 2021, 26, 2082 GLY143 (1.74 Å),~~~ 5 of 22 CYS145CYS145 (2.88 (2.88 Å),~~~ Å),~~~ CYS145 (2.88 Å),~~~ HIS163HIS163 (2.56 (2.56 Å),~~~ Å),~~~ SipholenolSipholenol H~~~ H~~~ HIS163 (2.56 Å),~~~ Sipholenol H~~~ S.S. siphonella siphonella Table 1. Cont. −12.0−12.0 HIS164HIS164 (2.85 (2.85 Å),~~~ Å),~~~ (157)(157) S. siphonella −12.0 HIS164 (2.85 Å),~~~ (157) MET165MET165 (2.74 (2.74 Å),~~~ Å),~~~ Docking MET165 (2.74 Å),~~~ Plant THR190 (1.74, 2.17 Å), a MNP Name 2D Chemical Structure Score THR190Binding (1.74, Features2.17 Å), Source THR190 (1.74, 2.17 Å), (kcal/mol) GLN192GLN192 (2.11 (2.11 Å) Å) GLN192 (2.11 Å) CYS145CYS145 (2.34 (2.34 Å),~~~ Å),~~~ CYS145CYS145 (2.34 Å),~~~ (2.34 Å), HIS41 (2.43 Å),~~~ HIS41 HIS41(2.43 Å),~~~ (2.43 Å), DahabinoneDahabinoneDahabinone AA~~~ A~~~ HIS41 (2.43 Å),~~~ Dahabinone A~~~ S. siphonellaS. siphonella −11.9 −11.9GLU166GLU166 (2.01, 2.23, (2.01, 2.38 2.23, Å),~~~ 2.38 Å), (162) S. siphonella −11.9 GLU166 (2.01, 2.23, 2.38 Å),~~~ (162)(162) S. siphonella −11.9 GLU166 (2.01,GLY143 2.23, (2.122.38 Å),Å),~~~ (162) GLY143GLY143 (2.12 (2.12 Å),~~~ Å),~~~ GLY143GLN189 (2.12 Å),~~~ (2.03 Å) GLN189 (2.03 Å) GLN189 (2.03 Å) GLN189 (2.03 Å) GLY143GLY143 (2.04 (2.04 Å),~~~ Å),~~~ GLY143 (2.04 Å),~~~ GLY143 (2.04 Å), CYS145CYS145 (2.89 (2.89 Å),~~~ Å),~~~ CYS145CYS145 (2.89 Å),~~~ (2.89 Å), SipholenolSipholenolSipholenol II~~~ I~~~ Sipholenol I~~~ S. siphonellaS.S. siphonella siphonella −11.8−11.8 −11.8 HIS163HIS163 HIS163(2.89 (2.89 Å),~~~ Å),~~~ (2.89 Å), (174) S. siphonella −11.8 HIS163 (2.89 Å),~~~ (174)(174) HIS164 (2.01 Å), (174) HIS164HIS164 (2.01 (2.01 Å),~~~ Å),~~~ HIS164GLU166 (2.01 (1.62, Å),~~~ 2.03, 2.52 Å) GLU166GLU166 (1.62, (1.62, 2.03, 2.03, 2.52 2.52 Å) Å) GLU166 (1.62, 2.03, 2.52 Å)

TYR54TYR54 (2.24, (2.24, 2.55 2.55 Å),~~~ Å),~~~ TYR54TYR54 (2.24, (2.24,2.55 Å),~~~ 2.55 Å), Lobophytosterol~~~LobophytosterolLobophytosterol~~~ ASN142ASN142ASN142 (1.91, (1.91, 2.75 (1.91,2.75 Å),~~~ Å),~~~ 2.75 Å), Lobophytosterol~~~L. depressumL.L. depressum depressum −11.5−11.5 −11.5 ASN142 (1.91, 2.75 Å),~~~ (188)(188)(188) L. depressum −11.5 GLU166GLU166GLU166 (1.86, (1.86, 2.66 (1.86,2.66 Å),~~~ Å),~~~ 2.66 Å), (188) GLU166 (1.86,ASP187 2.66 (2.94 Å),~~~ Å) ASP187ASP187 (2.94 (2.94 Å) Å) ASP187 (2.94 Å)

THR26THR26 (2.69 (2.69 Å),~~~ Å),~~~ (22R,24E,28E)-5(22R,24E,28E)-5β,6β,6,6β-β- THR26 (2.69 Å),~~~ (22R,24E,28E)-5β,6β- (22R,24E,28E)- HIS41HIS41 (2.14THR26 (2.14 Å),~~~ Å),~~~ (2.69 Å), Epoxy-22,28-oxido-24-Epoxy-22,28-oxido-24- HIS41 (2.14 Å),~~~ 5Epoxy-22,28-oxido-24-β,6β-Epoxy-22,28- CYS145CYS145 (2.45HIS41 (2.45 Å),~~~ (2.14Å),~~~ Å), methyl-5oxido-24-methyl-methyl-5ααcholestan-cholestan- L.L. depressum depressum −11.4−11.4 CYS145CYS145 (2.45 Å),~~~ (2.45 Å), methyl-5αcholestan-L. depressumL. depressum −11.4 −11.4 5αcholestan- ARG188ARG188 (2.03 (2.03 Å),~~~ Å),~~~ 3β3,25,28-triol~~~β,25,28-triol~~~,25,28-triol~~~ ARG188ARG188 (2.03 Å),~~~ (2.03 Å), MoleculesMolecules33ββ,25,28-triol ,25,28-triol~~~2021 2021, 26, 26, x, xFOR FOR PEER PEER REVIEW REVIEW THR190THR190 (2.06, 2.55 (2.06, Å),~~~ 2.55 Å),6 6of of 22 22 THR190 (2.06, 2.55 Å),~~~ (191)(191)(191) THR190 (2.06,GLN192 2.55 (2.26 Å),~~~ Å) (191) GLN192 (2.26 Å) GLN192 (2.26 Å) GLN192 (2.26 Å)

CYS44 (2.10 Å),~~~ CYS44 (2.10CYS44 Å),~~~ (2.10 Å), TasnemoxideTasnemoxideTasnemoxide AA~~~ A~~~ D.D.D. ery- ery- −11.4 −11.4 TYR54TYR54 (2.54, (2.54,2.97 Å),~~~ 2.97 Å), (144) erythraeanus −11.4 TYR54 (2.54, 2.97 Å),~~~ (144)(144) thraeanusthraeanus GLU166 (1.91, 1.97 Å) GLU166GLU166 (1.91, (1.91, 1.97 1.97 Å) Å)

GLY143GLY143 (1.95 (1.95 Å),~~~ Å),~~~ GLY143 (1.95 Å), Siphonellinol C~~~ GLN189 (2.09 Å),~~~ SiphonellinolSiphonellinol CC~~~ GLN189GLN189 (2.09 Å),~~~ (2.09 Å), S. siphonellaS.S. siphonella siphonella −11.3−11.3 −11.3 (172)(172)(172) THR190THR190THR190 (1.73, (1.73, 2.41 (1.73,2.41 Å),~~~ Å),~~~ 2.41 Å), GLN192 (2.10 Å) GLN192GLN192 (2.10 (2.10 Å) Å)

GLY143GLY143 (2.00 (2.00 Å),~~~ Å),~~~ Siphonellinol-C-23-hy-Siphonellinol-C-23-hy- GLN189GLN189 (1.97 (1.97 Å),~~~ Å),~~~ droperoxide~~~droperoxide~~~ S.S. siphonella siphonella −11.2−11.2 THR190THR190 (2.03, (2.03, 2.43 2.43 Å),~~~ Å),~~~ (171)(171) GLN192GLN192 (2.46 (2.46 Å) Å)

GLU166GLU166 (3.03 (3.03 Å),~~~ Å),~~~ ErylosidesErylosides K~~~ K~~~ ErylusErylus lenden- lenden- −11.1−11.1 HIS163HIS163 (2.18 (2.18 Å),~~~ Å),~~~ (224)(224) feldifeldi HIS164HIS164 (2.26, (2.26, 2.10 2.10 Å) Å)

THR190THR190 (2.08 (2.08 Å) Å),~~~,~~~ SipholenolSipholenol D~~~ D~~~ HIS163HIS163 (2.48 (2.48 Å),~~~ Å),~~~ S.S. siphonella siphonella −11.0−11.0 (176)(176) GLU166GLU166 (2.98 (2.98 Å),~~~ Å),~~~ GLN189GLN189 (1.91 (1.91 Å) Å)

Molecules 2021, 26, x FOR PEER REVIEW 6 of 22

Molecules 2021, 26, x FOR PEER REVIEW CYS44 (2.10 Å),~~~6 of 22 Tasnemoxide A~~~ D. ery- −11.4 TYR54 (2.54, 2.97 Å),~~~ (144) thraeanus CYS44 (2.10 Å),~~~ Tasnemoxide A~~~ D. ery- GLU166 (1.91, 1.97 Å) −11.4 TYR54 (2.54, 2.97 Å),~~~ (144) thraeanus CYS44 (2.10 Å),~~~ Tasnemoxide A~~~ D. ery- GLU166 (1.91, 1.97 Å) −11.4 TYR54 (2.54, 2.97 Å),~~~ (144) thraeanus GLU166 (1.91, 1.97 Å)

GLY143 (1.95 Å),~~~

MoleculesSiphonellinol2021, 26, 2082 C~~~ GLN189 (2.09 Å),~~~6 of 22 S. siphonella −11.3 GLY143 (1.95 Å),~~~ (172) THR190 (1.73, 2.41 Å),~~~ Siphonellinol C~~~ GLY143GLN189 (1.95 (2.09 Å),~~~ Å),~~~ S. siphonella −11.3 GLN192 (2.10 Å) Siphonellinol(172) C~~~ THR190GLN189 (2.09 (1.73, Å),~~~ 2.41 Å),~~~ S. siphonella Table 1. Cont. −11.3 (172) THR190GLN192 (1.73, 2.41 (2.10 Å),~~~ Å) Docking GLN192 (2.10 Å) Plant MNP Name 2D Chemical Structure Score Binding Features a Source (kcal/mol)

GLY143 (2.00 Å),~~~ Siphonellinol-C-23-hy- GLY143GLN189 (2.00 (1.97 Å),~~~ Å),~~~ Siphonellinol-C-23-hy-droperoxide~~~ S. siphonella −11.2 GLY143 (2.00GLY143 Å),~~~ (2.00 Å), Siphonellinol-C-23-Siphonellinol-C-23-hy- THR190 (2.03, 2.43 Å),~~~ GLN189GLN189 (1.97 (1.97Å),~~~ Å), hydroperoxidedroperoxide~~~(171) S. siphonellaS. siphonella −11.2 −11.2 GLN189 (1.97 Å),~~~ droperoxide~~~ S. siphonella −11.2 GLN192THR190 (2.03,(2.46 2.43Å) Å), (171) THR190THR190 (2.03, (2.03, 2.43 2.43 Å),~~~ Å),~~~ (171) (171) GLN192 (2.46 Å) GLN192GLN192 (2.46 (2.46 Å) Å)

GLU166 (3.03 Å),~~~ Erylosides K~~~ Erylus lenden- GLU166 (3.03GLU166 Å),~~~ (3.03 Å), ErylosidesErylosides K K~~~Erylus Erylus lenden- −11.1 GLU166HIS163 (3.03 (2.18 Å),~~~ Å),~~~ (224) feldi −11.1 −11.1 HIS163 (2.18HIS163 Å),~~~ (2.18 Å), Erylosides(224) K~~~ lendenfeldiErylus lenden- (224) feldi −11.1 HIS163HIS164HIS164 (2.18 (2.26, (2.26, Å),~~~ 2.10 2.10 Å) Å) (224) feldi HIS164 (2.26, 2.10 Å) HIS164 (2.26, 2.10 Å)

THR190THR190 (2.08 (2.08Å),~~~ Å) ,~~~ Sipholenol D~~~ HIS163 (2.48THR190 Å),~~~ (2.08 Å), SipholenolSipholenol D D~~~ S. siphonella −11.0 HIS163HIS163 (2.48 (2.48 Å),~~~, Å), S. siphonellaS. siphonella −11.0 −11.0 THR190 (2.08 Å) ~~~ (176)(176)(176) GLU166GLU166 (2.98GLU166 (2.98Å),~~~ (2.98 Å),~~~ Å), SipholenolMolecules 2021 D~~~,, 26,, x x FORFOR PEERPEER REVIEWREVIEW HIS163 (2.48 Å),~~~7 of 22 GLN189GLN189 (1.91 Å) (1.91 Å) S. siphonella −11.0 GLN189 (1.91 Å) (176) GLU166 (2.98 Å),~~~

GLN189 (1.91 Å)

HO OH GLY143 (1.93 Å),~~~ GLY143 (1.93 Å), SipholenoneSipholenone A A~~~ HIS163 (2.41HIS163 Å),~~~ (2.41 Å), S. siphonellaS. siphonella −11.0 −11.0 (175) (175) HIS164 (2.54HIS164 Å),~~~ (2.54 Å), GLU166 (2.36 Å) O GLU166 (2.36 Å)

HO OH

ASN142 (2.26 Å),~~~ ASN142 (2.26 Å), NeviotineNeviotine B B~~~ GLU166 (1.95GLU166 Å),~~~ (1.95 Å), S. siphonellaS. siphonella −10.9 −10.9 (158) (158) GLN189 (1.80GLN189 Å),~~~ (1.80 Å), O THR190 (2.57 Å) HO THR190 (2.57 Å)

O OH

ASN142 (2.32 Å),~~~ Eryloside A~~~ HO Genus Erylus −10.7 GLU166 (1.95 Å),~~~ (197) O THR190 (1.87 Å)

GLN189 (1.77 Å),~~~ Sipholenone D~~~ S. siphonella −10.7 THR190 (1.83 Å),~~~ (155) GLN192 (2.23 Å)

24-Methylcholestane-5- MET49 (2.16 Å),~~~ en-3β,25-diol~~~,25-diol~~~ S. polydactyla −10.6 PHE140 (1.87 Å),~~~ (187) GLN189 (2.97 Å)

SipholenolA-4-O-3′′,4,4′′- HIS163 (2.30 Å),~~~ dichlorobenzoate~~~ S. siphonella −10.5 GLN189 (1.78 Å) (151)

Molecules 2021, 26, x FOR PEER REVIEW 7 of 22 Molecules 2021, 26, x FOR PEER REVIEW 7 of 22 Molecules 2021, 26, x FOR PEER REVIEW 7 of 22

HO HO OH HO OH HO GLY143 (1.93 Å),~~~ OH GLY143 (1.93 Å),~~~ OH Sipholenone A~~~ GGLY143HIS163LY143 (2.41 (1.93 Å),~~~ Å),~~~ Sipholenone A~~~ S. siphonella −11.0 HIS163 (2.41 Å),~~~ Sipholenone(175) A~~~ S. siphonella −11.0 HIS164HIS163 (2.54(2.41 Å),~~~ Sipholenone(175) A~~~ S. siphonella −11.0 HIS164HIS163 (2.54(2.41 Å),~~~ (175) S. siphonella −11.0 HIS164 (2.54 Å),~~~ (175) O HIS164GLU166 (2.54 (2.36 Å),~~~ Å) O GLU166 (2.36 Å) O GLU166 (2.36 Å) O GLU166 (2.36 Å) O O O O HO HO OH HO OH HO OH OH Molecules 2021, 26, 2082 ASN142ASN142 (2.26 Å),~~~7 of 22 Neviotine B~~~ GLU166ASN14ASN1422 (1.95(2.26 Å),~~~ Neviotine B~~~ S. siphonella −10.9 GLU166 (1.95 Å),~~~ Neviotine(158) B~~~ S. siphonella −10.9 GLN189GLU166 (1.80(1.95 Å),~~~ (158) S. sisiphonellaphonella −1010.9.9 GLN189 (1.80 Å),~~~ (158) O GLN189 (1.80 Å),~~~ (158) HO TableO 1. Cont. GLN189THR190 (1.80 (2.57 Å),~~~ Å) HO THR190 (2.57 Å) O HO O THR190 (2.57 Å) HO Docking THR190 (2.57 Å) Plant a MNP Name O2D Chemical Structure Score Binding Features O Source OH O OH (kcal/mol) O OH OH HO HO HO HO ASN142 (2.32 Å),~~~ HO ASN14ASN1422 (2.32 (2.32 Å),~~~ Å), ErylosideEryloside A A~~~ HO Eryloside A~~~ GenusGenus Erylus Erylus −10.7 −10.7 GLU166ASN14GLU1662 (1.95(2.32 (1.95 Å),~~~ Å), Genus Erylus HO O −10.7 GLU166ASN142 (1.95(2.32 Å),~~~ Eryloside(197)(197) A~~~ HO O Eryloside(197) A~~~ Genus Erylus −10.7 GLU166THR190 (1.95 (1.87 Å),~~~ Å) Genus Erylus O −10.7 GLU166THR190 (1.95 (1.87 Å),~~~ Å) (197) O THR190 (1.87 Å) THR190 (1.87 Å) O O O O

GLN189 (1.77 Å),~~~ Sipholenone D~~~ GLN189 (1.77 Å), SipholenoneSipholenone D D~~~ S. siphonella −10.7 THR190GLN189 (1.83(1.77 Å),~~~Å),~~~ S. siphonellaS. siphonella −10.7 −10.7 THR190GLN189THR190 (1.83(1.77 (1.83 Å),~~~Å),~~~ Å), Sipholenone(155)(155) D~~~ (155) S. siphonella −10.7 THR190GLN192GLN192 (1.83 (2.23 (2.23 Å),~~~ Å) Å) (155) S. siphonella −10.7 THR190GLN192 (1.83 (2.23 Å),~~~ Å) (155) GLN192 (2.23 Å) GLN192 (2.23 Å)

24-Methylcholestane-5- MET49MET49 (2.16 Å),~~~ 24-Methylcholestane-5- MET49 (2.16 Å),~~~ 24-Methylcholestane-5-24-Methylcholestane-5-en-3β,25-diol~~~ S. polydactyla −10.6 PHE140MET49MET49 (2.16(1.87 (2.16 Å),~~~ Å), en-3β,25-diol~~~ S.S. polydactyla −10.6 PHE140 (1.87 Å),~~~ en-3en-3β,25-diolβ,25-diol~~~ S. polydactyla −10.6 −10.6 PHE140PHE140 (1.87 (1.87 Å),~~~ Å), en-3β(187),25-diol~~~ polydactylaS. polydactyla −10.6 PHE140GLN189 (1.87 (2.97 Å),~~~ Å) (187)(187) GLN189GLN189 (2.97 (2.97 Å) Å) (187) GLN189 (2.97 Å)

SipholenolA-4-O-3′,4′- SipholenolA-4-O-3′,4′- HIS163 (2.30 Å),~~~ SipholenolA-4-O-3dichlorobenzoate~~~′,4′ - S. siphonella −10.5 HIS163 (2.30 Å),~~~ SipholenolA-4-O-3dichlorobenzoate~~~0,40- S. siphonella −10.5 HIS163 (2.30 Å),~~~ HIS163GLN189HIS163 (2.30 (1.78 (2.30 Å),~~~ Å) Å), dichlorobenzoatedichlorobenzoate~~~(151) S. siphonellaS. sisiphonellaphonella −1010.5.5 −10.5 GLN189 (1.78 Å) (151) GLN189GLN189 (1.78 (1.78 Å) Å) (151)(151) GLN189 (1.78 Å) Molecules 2021(151), 26, x FOR PEER REVIEW 8 of 22

Molecules 2021, 26, x FOR PEER REVIEW 8 of 22

StigmasterolStigmasterol~~~ MET49MET49 (2.18 (2.18 Å),~~~ Å), Stigmasterol~~~D. coccineaD. coccinea −10.5 −10.5 MET49 (2.18 Å),~~~ (220)(220) D. coccinea −10.5 GLN189GLN189 (2.97 (2.97 Å) Å) (220) GLN189 (2.97 Å)

Cholest-5-en-3Cholest-5-en-3Cholest-5-en-3β,7β- β,7ββ-,7β- MET49 (2.17MET49 Å),~~~ (2.17 Å), diol diol~~~ A. dichotomaA. dichotoma −10.3 −10.3 MET49 (2.17 Å),~~~ diol~~~ A. dichotoma −10.3 GLN189GLN189 (1.94 Å) (1.94 Å) (206) (206) GLN189 (1.94 Å) (206)

Campesterol~~~ MET49 (2.17 Å),~~~ D. coccinea −10.3 Campesterol~~~(221) GLN189MET49 (3.01 (2.17 Å) Å),~~~ D. coccinea −10.3 (221) GLN189 (3.01 Å)

Cholesterol~~~ Dendronephthy MET49 (2.10 Å),~~~ −10.3 (184) a GLN189 (2.97 Å) Cholesterol~~~ Dendronephthy MET49 (2.10 Å),~~~ −10.3 (184) a GLN189 (2.97 Å)

Clionasterol~~~ Dragmacidon MET49 (2.16 Å),~~~ −10.3 (219) coccinea GLN189 (2.95 Å)

Brassicasterol~~~ D. coccinea −10.1 MET49 (2.17 Å) (222)

Molecules 2021, 26, x FOR PEER REVIEW 8 of 22 Molecules 2021,, 26,, xx FORFOR PEERPEER REVIEWREVIEW 8 of 22

Molecules 2021, 26, x FOR PEER REVIEW 8 of 22

Stigmasterol~~~ MET49 (2.18 Å),~~~ D. coccinea −10.5 (220) GLN189 (2.97 Å)

Stigmasterol~~~ MET49 (2.18 Å),~~~ D. coccinea −10.5 (220) GLN189 (2.97 Å) Molecules 2021, 26, 2082 8 of 22

Cholest-5-en-3β,7,7β- MET49 (2.17 Å),~~~ diol~~~ A. dichotoma −10.3 Table 1. Cont. GLN189 (1.94 Å) (206) Cholest-5-en-3β,7β- Docking Plant MNP Name 2D Chemical Structure Score MET49 Binding(2.17 Å),~~~ Features a diol~~~ SourceA. dichotoma −10.3 (kcal/mol)GLN189 (1.94 Å) (206)

CampesterolCampesterol~~~ MET49MET49 (2.17 (2.17 Å),~~~ Å), Campesterol~~~ D. coccineaD. coccinea −10.3 −10.3 MET49 (2.17 Å),~~~ (221) D. coccinea −10.3 GLN189 (3.01 Å) (221) GLN189 (3.01 Å) Campesterol~~~ MET49 (2.17 Å),~~~ D. coccinea −10.3 (221) GLN189 (3.01 Å)

Cholesterol~~~ Dendronephthy MET49 (2.10 Å),~~~ CholesterolCholesterol~~~ Dendronephthy MET49MET49 (2.10 (2.10 Å),~~~ Å), Dendronephthya −10.3 −10.3 (184)(184) a GLN189GLN189 (2.97 (2.97 Å) Å) (184)Cholesterol~~~ Dendronephthya MET49GLN189 (2.10 Å),~~~ (2.97 Å) −10.3 (184) a GLN189 (2.97 Å)

Clionasterol~~~ Dragmacidon MET49 (2.16 Å),~~~ ClionasterolClionasterol~~~ DragmacidonDragmacidon MET49MET49 (2.16 (2.16 Å),~~~ Å), Clionasterol~~~ Dragmacidon −10.3 −10.3 MET49 (2.16 Å),~~~ (219)(219) coccineacoccinea −10.3 GLN189GLN189 (2.95 (2.95 Å) Å) (219) coccinea GLN189 (2.95 Å)

BrassicasterolBrassicasterol~~~Brassicasterol~~~ Molecules 2021, 26, x FOR PEERD. coccinea REVIEWD. coccineaD. coccinea −10.1−10.1 −10.1 MET49 MET49MET49 (2.17 (2.17Å) (2.17 Å) Å) 9 of 22 (222) D. coccinea −10.1 MET49 (2.17 Å) (222)(222)

3β- 3β- Hexadecanoylcholest-Hexadecanoylcholest-5- A. dichotomaA. dichotoma −10.0 −10.0 GLY143GLY143 (1.95 (1.95 Å) Å) 5-en-7-oneen-7-one~~~ (202) (202)

Sipholenone E~~~ S. siphonella −9.9 GLN189 (1.83 Å) (163)

a Only hydrogen bonds (in Å) were listed. b No plant source was noticed.

Figure 1. 2D representations of the predicted binding modes of MNPs (i) 190, (ii) 178, (iii) 226, and (iv) lopinavir towards SARS-CoV-2 main protease (Mpro).

MoleculesMolecules 2021, 262021, x ,FOR 26, x PEER FOR PEERREVIEW REVIEW 9 of 229 of 22

3β- 3β- Hexadecanoylcholest-5-MoleculesHexadecanoylcholest-5-2021, 26, 2082 9 of 22 A. dichotomaA. dichotoma −10.0− 10.0 GLY143 GLY143 (1.95 (1.95 Å) Å) en-7-one~~~en-7-one~~~ (202)(202) Table 1. Cont.

Docking Plant MNP Name 2D Chemical Structure Score Binding Features a Source (kcal/mol)

Sipholenone E~~~ SipholenoneSipholenone E E~~~ S. siphonellasiphonellaS. siphonella −9.9 −9.9 −9.9 GLN189 GLN189GLN189 (1.83 (1.83 Å) (1.83 Å) Å) (163)(163)

a b Onlya Onlya Onlyhydrogen hydrogen hydrogen bonds bonds bonds (in (in Å) Å) (in were were Å) were listed.listed. listed.b NoNo plant plantb No source plant source wassource was noticed. noticed. was noticed.

FigureFigure 1. 2D 1.representations 2D representations of the of pred predicted the ictedpred ictedbinding binding modes modes of MNPs of MNPs ( i) 190, (i) ( 190,ii) 178, (ii) ( 178,iii) 226, (iii) and 226, ( andiv) lopinavir (iv) lopinavir towards towards pro SARS-CoV-2SARS-CoV-2 main main protease protease (M(Mpro ).(M). pro).

Erylosides B (226) isolated from Erylus lendenfeldi also exhibited a high binding afﬁnity against SARS-CoV-2 Mpro with a docking score of −12.1 kcal/mol. in silico Mpro binding in the active site indicated that the ﬁve carbonyl groups of methyl acetates exhibit seven hydrogen bonds with a backbone OH of TYR54, imidazole ring of HIS41, SH of CYS145, imidazole ring of HIS163, NH of ASN142, NH of GLY143, and NH2 of GLN189 with bond lengths of 3.05, 1.96, 2.04, 1.93, 1.75, 2.82, and 1.91 Å, respectively (Figure1 and Table1). Moreover, the oxygen of the (cyclohexyloxy) cyclohexane ring forms a hydrogen bond with the backbone NH of GLU166 with a bond length of 2.37 Å (Figure1 and Table1). The Molecules 2021, 26, 2082 10 of 22

oxygen atom of the methoxy group is also hydrogen bound to the NH of ASN142 with a bond length of 2.96 Å (Figure1 and Table1). Compared with 190, 178, and 226, lopinavir presented a similar binding afﬁnity towards Mpro with a docking score of −9.8 kcal/mol. Scrutinizing the binding mode of lopinavir inside the active site of Mpro revealed that the NH of the tetrahydro-1-methylpyrimidin-2(1H)-one ring exhibited two hydrogen bonds with a backbone hydroxy group of SER144 and a carbonyl group of LEU141 with bond lengths of 3.09 and 1.96 Å, respectively. The carbonyl group of the tetrahydro-1-methylpyrimidin-2(1H)-one ring was also observed to form a hydrogen bond with the backbone NH of GLY143 with a bond length of 2.01 Å, and the NH of the methylacetamide group participates in a hydrogen bond with the carbonyl group of HIS164 with a bond length of 2.62 Å (Figure1 and Table1). The current results provide quantitative data of the binding afﬁnities of 190, 178, 226, and lopinavir as promising SARS-CoV-2 Mpro inhibitors.

2.2. Molecular Dynamics (MD) Simulations MD simulations probe the stability of the ligand–enzyme complexes, conformational flexibilities, structural details, and the dependability of ligand–enzyme affinities [24,25]. Therefore, those natural products with low Mpro docking scores were submitted for MD simulations followed by binding energy calculations. The simulations were conducted with an implicit water solvent for 250 ps, and the molecular mechanics/generalized Born surface area (MM/GBSA) approach was applied to estimate the corresponding binding affinities; thereby diminishing the time and computational costs (see computational methodology section for details). The calculated MM/GBSA binding energies for the pre-screened natural products are summarized in Table S2. Six compounds had lower binding energies (∆Gbinding) than lopinavir (calc. −39.4 kcal/mol). These metabolites were further subjected to a 10 ns MD simulation in an explicit water solvent to obtain more accurate Mpro binding affinities. MD/MM/GBSA binding energies were estimated (Figure2). Four of these compounds exhibited lower binding energies (∆Gbinding) compared to lopinavir (calc. −35.4 kcal/mol). Moreover, those potent MNPs were chosen and submitted for 50 ns MD simulations in the explicit water solvent, and the corresponding binding energies were evaluated (Figure2). Only erylosides B (226) exhibited steady binding throughout the simulation, while 202, 151, and 224 showed an increase in MM/GBSA binding energies over the simulated time. For instance, the calculated MM/GBSA binding energies for 226 against Mpro were −50.8, −48.8 and −50.0 kcal/mol over 250 ps implicit-solvent MD, 10 ns explicit-solvent MD, and 50 ns explicit-solvent MD simulations, respectively. This shows the importance of long MD simulations to predict metabolite-Mpro binding affinities. Therefore, MD simulations for the 226-Mpro complex were prolonged to 100 ns, and the corresponding MM/GBSA binding energy was calculated (Figure2). No appreciable variance between the calculated MM/GBSA binding energy for the 226-Mpro complex throughout the 50 ns MD simulation and the corresponding MM/GBSA binding energy throughout the 100 ns MD simulation were observed (Figure2). Compared with lopinavir, 226 showed a higher binding affinity towards Mpro over the 100 ns MD simulation with an average ∆Gbinding of −51.9 kcal/mol. Hydrogen bonding, pi-based, hydrophobic, and van der Waals interactions with essential amino acid residues within the Mpro active site are all thought to contribute to the strong binding constant. The 2D and 3D representations for 226 and lopinavir within the Mpro active site over 100 ns are shown in Figure3 and S3, respectively. Interestingly, 226 maintained hydrogen bonding with the proximal amino acid residues of Mpro over the 100 ns MD course (Figure3). Structural insights into the binding mode of the 226 with the Mpro demonstrated that the carbonyl groups of methyl acetates form four hydrogen bonds with the backbone imidazole of the HIS41, OH of SER146, the carbonyl group of ASN142, and NH2 of GLN189, with bond lengths of 2.72, 2.76, 3.37, and 1.89 Å, respectively (Figure3); while the oxygen of the (cyclohexyloxy) cyclohexane ring forms hydrogen bonds with the backbone NH2 of Molecules 2021, 26, x FOR PEER REVIEW 10 of 22

The carbonyl group of the tetrahydro-1-methylpyrimidin-2(1H)-one ring was also observed to form a hydrogen bond with the backbone NH of GLY143 with a bond length of 2.01 Å, and the NH of the methylacetamide group participates in a hydrogen bond with the carbonyl group of HIS164 with a bond length of 2.62 Å (Figure 1 and Table 1). The current results provide quantitative data of the binding affinities of 190, 178, 226, and lopinavir as promising SARS-CoV-2 Mpro inhibitors.

2.2. Molecular Dynamics (MD) Simulations MD simulations probe the stability of the ligand–enzyme complexes, conformational flexibilities, structural details, and the dependability of ligand–enzyme affinities [24,25]. Therefore, those natural products with low Mpro docking scores were submitted for MD simulations followed by binding energy calculations. The simulations were conducted with an implicit water solvent for 250 ps, and the molecular mechanics/generalized Born surface area (MM/GBSA) approach was applied to estimate the corresponding binding affinities; thereby diminishing the time and computational costs (see computational methodology section for details). The calculated MM/GBSA binding energies for the pre- screened natural products are summarized in Table S2. Six compounds had lower binding energies (ΔGbinding) than lopinavir (calc. −39.4 kcal/mol). These metabolites were further subjected to a 10 ns MD simulation in an explicit water solvent to obtain more accurate Mpro binding affinities. MD/MM/GBSA binding energies were estimated (Figure 2). Four of these compounds exhibited lower binding energies (ΔGbinding) compared to lopinavir (calc. −35.4 kcal/mol). Moreover, those potent MNPs were chosen and submitted for 50 ns MD simulations in the explicit water solvent, and the corresponding binding energies were evaluated (Figure 2). Only erylosides B (226) exhibited steady binding throughout the simulation, while 202, 151, and 224 showed an increase in MM/GBSA binding energies Molecules 2021, 26, 2082 over the simulated time. For instance, the calculated MM/GBSA binding energies for 226 11 of 22 against Mpro were −50.8, −48.8 and −50.0 kcal/mol over 250 ps implicit-solvent MD, 10 ns explicit-solvent MD, and 50 ns explicit-solvent MD simulations, respectively. This shows Molecules 2021, 26, x FOR PEER theREVIEW importance of long MD simulations to predict metabolite-Mpro binding affinities.11 of 22

Therefore,GLN189 MD simulations with a bond for the length 226-M ofpro 2.62 complex Å (Figure were3 prolonged). Furthermore, to 100 thens, oxygenand the of tetrahydro-2H- correspondingpyran MM/GBSA bonds with binding the backbone energy was NH2 calculated of GLN189 (Figure with 2). a bond length of 2.06 Å (Figure3). No appreciable variance between the calculated MM/GBSA binding energy for the 226-Mpro complex throughout the 50 ns MD simulation and the corresponding MM/GBSA binding energy throughout the 100 ns MD simulation were observed (Figure 2). Compared with lopinavir, 226 showed a higher binding affinity towards Mpro over the 100 ns MD simulation with an average ΔGbinding of −51.9 kcal/mol. Hydrogen bonding, pi- based, hydrophobic, and van der Waals interactions with essential amino acid residues within the Mpro active site are all thought to contribute to the strong binding constant. The 2D and 3D representations for 226 and lopinavir within the Mpro active site over 100 ns are shown in Figure 3 and S3, respectively. Interestingly, 226 maintained hydrogen bonding with the proximal amino acid residues of Mpro over the 100 ns MD course (Figure 3). Structural insights into the binding mode of the 226 with the Mpro demonstrated that the carbonyl groups of methyl acetates form four hydrogen bonds with the backbone imidazole of the HIS41, OH of SER146, the carbonyl group of ASN142, and NH2 of GLN189, with bond lengths of 2.72, 2.76, 3.37, and 1.89 Å, respectively (Figure 3); while the oxygen of the (cyclohexyloxy) cyclohexane ring forms hydrogen bonds with the backbone NH2 of GLN189 with a bond length of 2.62 Å (Figure 3). Furthermore, the oxygen of tetrahydro-2H-pyran bonds with the backbone NH2 of GLN189 with a bond length of 2.06 Å (Figure 3).

While lopinavir also displayed high Mpro binding energy over the 100 ns MD Figure 2. AverageFigure molecularsimulation 2. Average mechanics/generalized at molecular ΔGbinding mechanics/generalized−33.6 kcal/mol, Born surface only three Born area surface (MM/GBSA)hydrogen area bonds (MM/GBSA) binding were energies observedbinding for with lopinavir the and the pro top natural productsenergies complexedproximal for lopinavir amino with Mand acidpro the overresidues top 250natural nsof in Mproducts anpro (Figure implicit complexed 3). water A comparisonwith solvent, M andover of 10250 the ns, ns erylosides 50in an ns, implicit and B 100 (226) ns molecular water solvent, and 10 ns, 50 ns, and 100 ns molecular dynamics (MD) simulations in an explicit and lopinavir revealed that the binding affinity of erylosides B (226) was approximately dynamics (MD) simulationswater solvent. in an explicit water solvent. two times higher than that of lopinavir.

Figure 3. 2D representationsFigure 3. of 2D binding representations modes of of (bindingi) erylosides modes B of (226)- (i) erylosides and (ii) B lopinavir-M (226)- and (iipro) lopinavir-Mcomplexespro according to an average structure overcomplexes a 100 ns MDaccording simulation. to an average structure over a 100 ns MD simulation.

Molecules 2021, 26, 2082 12 of 22

Molecules 2021, 26, x FOR PEER REVIEW 12 of 22 While lopinavir also displayed high Mpro binding energy over the 100 ns MD simulation at ∆Gbinding −33.6 kcal/mol, only three hydrogen bonds were observed with the proximal amino acid residues of Mpro (Figure3). A comparison of the erylosides B (226) The estimated MM/GBSA binding energies were further decomposed into separate and lopinavir revealed that the binding affinity of erylosides B (226) was approximately components to indicate the force in the binding of Mpro with erylosides B (226) and two times higher than that of lopinavir. lopinavir (Table 2). Evdw was a significant contributor to the erylosides B (226)- and The estimated MM/GBSA binding energies were further decomposed into separate lopinavir-Mpro binding affinities with average values of −71.2 and −45.6 kcal/mol, components to indicate the force in the binding of Mpro with erylosides B (226) and lopinavir respectively. Eele was efficient with an average value of −30.5 and −22.1 kcal/mol for the pro (Table2). Evdw was a significant contributor to the erylosides B (226)- and lopinavir-M erylosides B (226)- and lopinavir-Mpro binding affinities, respectively. binding affinities with average values of −71.2 and −45.6 kcal/mol, respectively. Eele was efficient with an average value of −30.5 and −22.1 kcal/mol for the erylosides B (226)- and Table 2. Components of the MM/GBSA binding energies for erylosides B (226)- and lopinavir-Mpro lopinavir-Mpro binding affinities, respectively. complexes as determined by MD simulation at 100 ns.

Table 2. ComponentsCompound of the MM/GBSA bindingCalculated energies for MM erylosides/GBSA B Binding (226)- and Energy lopinavir-M (kcal/mol)pro complexes as determined by MD simulation atName 100 ns. ∆EVDW a ∆Eele b ∆EGB c ∆ESUR d ∆Ggas e ∆GSolv f ∆Gbinding g Erylosides B −71.2 Calculated−30.5 MM/GBSA58.1 Binding−8.3 Energy−101.7 (kcal/mol) 49.8 −51.9 Compound Name (226) a b c d e f g Lopinavir∆EVDW −45.6∆Eele −22.1∆ EGB 39.9 ∆ESUR−5.7 ∆G−67.8gas 34.2∆G Solv −33.6 ∆Gbinding a b c Erylosides B (226) van der− Waals71.2 energy.− 30.5electrostatic energy. 58.1 The electrostatic−8.3 solvation−101.7 free energy 49.8calculated −51.9 Lopinavir from the− generalized45.6 Born−22.1 equation. d The 39.9 nonpolar component−5.7 of the− 67.8solvation energy. 34.2 e Total gas −33.6 phase energy. f The solvation free energy. g The evaluated free energy calculated from the terms a van der Waals energy. b electrostatic energy. c The electrostatic solvation free energy calculated from the generalized Born equation. d The nonpolar component of the solvationabove. energy. e Total gas phase energy. f The solvation free energy. g The evaluated free energy calculated from the terms above.

The bindingThe energies binding of energies erylosides of erylosides B (226)- Band (226)- lopinavir-M and lopinavir-Mpro complexespro complexes were were further further decomposeddecomposed at the at the per-residue per-residue level level and and the residuesresidues with with absolute absolute energy energy contribution < contribution− <0.50 −0.50 kcal/mol kcal/mol were were shown shown (Figure (Figure4). MET165,4). MET165, GLU166, GLU166, PRO168, PRO168, and and GLN189 in the GLN189 in theMpro Mprocomplex complex favorably favorably participate participate with with erylosides erylosides B B (226) (226) and and lopinavir. lopinavir. There was a There was aconsiderable considerable contribution contribution by by GLN189 GLN189 to to the the total total binding binding free free energy energy with with values of −4.6 values of −4.6and and−3.3 − kcal/mol3.3 kcal/mol for erylosidesfor erylosides B (226) B (226) and lopinavir,and lopinavir, respectively. respectively. Additionally, the Additionally,hydrophobic the hydrophobic residues residues share greatershare gr energieseater energies as a result as a ofresult the formation of the formation of the hydrophobic of the hydrophobicinteractions interactions between between erylosides erylosides B (226) andB (226) hydrophobic and hydrophobic residues. residues.

Figure 4. EnergyFigure contributions 4. Energy contributions (kcal/mol) for (kcal/mol) Mpro amino for acidMpro residuesamino acid to residues the binding to the free binding energy free of erylosides energy B (226) and lopinavir.of erylosides B (226) and lopinavir.

2.3. Post-Dynamics Analyses 2.3. Post-Dynamics Analyses To further confirmed the stability and behavior of erylosides B (226) complexed with Mpro, structural andTo furtherenergetic conﬁrmed analysesthe were stability executed and during behavior 100 of ns erylosides MD simulations B (226) complexedand with pro compared toM those, structural of lopinavir. and energeticControl of analyses the structural were executed stability duringof the 100studied ns MD system simulations and was accomplished by investigating root-mean-square deviation (RMSD), center-of-mass (CoM) distance, hydrogen bond length, and binding energy per frame.

Molecules 2021, 26, 2082 13 of 22

Molecules 2021, 26, x FOR PEER REVIEWcompared to those of lopinavir. Control of the structural stability of the studied system13 of 22

was accomplished by investigating root-mean-square deviation (RMSD), center-of-mass (CoM) distance, hydrogen bond length, and binding energy per frame.

2.3.1. Binding Energy per Frame The globalglobal structuralstructural stabilitystability ofof erylosideserylosides BB (226)(226) andand lopinavirlopinavir inin complexcomplex withwith pro Mpro waswas estimated estimated over over the the 100 100 ns ns MD MD simula simulationstions by measuring the correlation betweenbetween thethe bindingbinding energyenergy perper frameframe andand timetime (Figure(Figure5 5).). Overall Overall stabilities stabilities for for erylosides erylosides B B (226) (226) and lopinavir exhibited average binding energies (ΔGbinding) of −51.9 and −33.6 kcal/mol, and lopinavir exhibited average binding energies (∆Gbinding) of −51.9 and −33.6 kcal/mol, respectively.respectively. AccordingAccording toto thisthis analysis,analysis, allall complexescomplexes conservedconserved stabilitystability throughoutthroughout thethe 100100 nsns MDMD simulation.simulation.

Figure 5. Calculated MM/GBSA MM/GBSA binding binding energy energy per per fram framee for for erylosides erylosides B (black) B (black) and and lopinavir lopinavir (red)(red) withwith MMpropro overover 100 100 ns MD simulations.

2.3.2. Hydrogen Bond Length pro Hydrogen bond analyses for eryl erylosidesosides B B (226)- (226)- and and lopinavir-M lopinavir-Mpro complexescomplexes over over a a100 100 ns ns MD MD simulation simulation were were performed performed (Tab (Tablele 33).). FourFour hydrogenhydrogen bondsbonds withwith GLN189,GLN189, pro GLU166, CYS145,CYS145, andand ASN142ASN142 werewere observedobserved withinwithin thethe erylosideserylosides BB (226)-M(226)-Mpro complex. AverageAverage bondbond lengthslengths werewere 2.9,2.9, 2.8,2.8, 2.9,2.9, andand 2.72.7 Å,Å, withwith occupationoccupation percentages of 96, 92, 91, andand 83%, 83%, respectively. respectively. In contrast,In contrast, two hydrogentwo hydrogen bonds werebonds detected were detected for the lopinavir- for the pro Mlopinavir-Mcomplex,pro withcomplex, GLN189 with and GLN189 GLY143 and showing GLY143 an showing average bond an average length ofbond 2.8 andlength 2.7 Åof and2.8 and an occupation 2.7 Å and an percentage occupation of 85.6percentage and 75.6%, of 85.6 respectively. and 75.6%, Overall, respectively. these hydrogen Overall, bondthese analyses supported the ﬁnding of high stability for the erylosides B (226)-Mpro complex hydrogen bond analyses supported the finding of high stability for the erylosides B (226)- compared to lopinavir-Mpro. Mpro complex compared to lopinavir-Mpro. 2.3.3. Center-of-Mass Distance Table 3. Distance, occupancy, and hydrogen bonding for erylosides B (226) and lopinavir with key Mpro amino acid pro residues. To gain a more in-depth insight into the stability of ligand-M during the MD simulation, center-of-mass (CoM) distances were measured (Figure6). Interestingly, CoM Compound Name Acceptodistancesr were consistentDono forr the erylosidesDistance B (226)-M (Å)pro a complexAngle (degree) compared a Occupied to lopinavir- (%) b GLN_189@OMpro, with averageErylosides values B of @O5-H29 5.6 vs. 5.9 Å, respectively. 2.9 This suggests 142 that erylosides 95.7 B Erylosides B~~~ GLU166@O(226) bound moreErylosides tightly to B the@O3-H16 Mpro complex than2.8 lopinavir. 141 92.3 (226) CYS145@O Erylosides B @O12-H44 2.9 152 91.1 ASN142@O Erylosides B @O16-H29 2.7 156 83.3 GLN189@O Lopinavir @O9-H19 2.8 145 85.6 Lopinavir GLY143@O Lopinavir @O12-H28 2.7 158 75.6 a The hydrogen bonds are inspected by the acceptor-H-donor angle of >120° and acceptor-donor atom distance < 3.5 Å. b Occupancy is employed to estimate the strength and stability of the hydrogen bond.

Molecules 2021, 26, 2082 14 of 22

Table 3. Distance, occupancy, and hydrogen bonding for erylosides B (226) and lopinavir with key Mpro amino acid residues.

Compound Angle Occupied Acceptor Donor Distance (Å) a Name (degree) a (%) b Erylosides B GLN_189@O 2.9 142 95.7 Molecules 2021, 26, x FOR PEER REVIEW @O5-H29 14 of 22 Erylosides B GLU166@O 2.8 141 92.3 Erylosides B @O3-H16 Erylosides B (226) CYS145@O 2.9 152 91.1 @O12-H44 2.3.3. Center-of-Mass DistanceErylosides B ASN142@O 2.7 156 83.3 @O16-H29 To gain a more in-depth insight into the stability of ligand-Mpro during the MD Lopinavir GLN189@O 2.8 145 85.6 simulation, center-of-mass (CoM)@O9-H19 distances were measured (Figure 6). Interestingly, CoM Lopinavir distances were consistent foLopinavirr the erylosides B (226)-Mpro complex compared to lopinavir- GLY143@O 2.7 158 75.6 Mpro, with average values of@O12-H28 5.6 vs. 5.9 Å, respectively. This suggests that erylosides B a The hydrogen bonds are inspected by the acceptor-H-donor angle of >120◦ and acceptor-donor atom pro (226)distance bound < 3.5 Å. b moreOccupancy tightly is employed to the to estimate M complex the strength andthan stability lopinavir. of the hydrogen bond.

FigureFigure 6. 6.Center-of-mass Center-of-mass (CoM) (CoM) distances distances (in Å) between(in Å) betwee erylosidesn erylosides B (black) and B lopinavir(black) and (red) lopinavir (red) pro andand GLN189GLN189 of of M Mproover over a 100 a ns100 MD ns simulation. MD simulation. 2.3.4. Root-Mean-Square Deviation 2.3.4.To Root-Mean-Square watch the structural stabilityDeviation of the erylosides B (226)- and lopinavir-Mpro complexes,To the watch root-mean-square the structural deviation stability (RMSD) valuesof th ofe theerylosides backbone atomsB (226)- of the wholeand lopinavir-Mpro complexes,complex were the estimated root-mean-square (Figure7). Unambiguously, deviation (R theMSD) estimated values RMSD of valuesthe backbone for the atoms of the investigated complexes remained below 0.20 nm over the 100 ns MD simulations. The wholeerylosides complex B (226)- andwere lopinavir-M estimatedpro complexes(Figure 7). reached Unambiguously, the steady state the in the estimated ﬁrst 10 ns RMSD values forMDs, the and investigated continued in complexes an almost stationary remained state below until the0.20 end nm of over the simulations. the 100 ns The MD simulations. Thecurrent erylosides results emphasized B (226)- that and the lopinavir-M erylosides B (226)pro complexes is tightly bonded reached and does the notsteady impact state in the first pro 10the ns overall MDs, topology and continued of SARS-CoV-2 in an M almost. stationary state until the end of the simulations. The current results emphasized that the erylosides B (226) is tightly bonded and does not impact the overall topology of SARS-CoV-2 Mpro.

Figure 7. Root-mean-square deviation (RMSD) of the backbone atoms from the initial structure of erylosides B (black) and lopinavir (red) with Mpro throughout a 100 ns MD simulation.

Molecules 2021, 26, x FOR PEER REVIEW 14 of 22

2.3.3. Center-of-Mass Distance To gain a more in-depth insight into the stability of ligand-Mpro during the MD simulation, center-of-mass (CoM) distances were measured (Figure 6). Interestingly, CoM distances were consistent for the erylosides B (226)-Mpro complex compared to lopinavir- Mpro, with average values of 5.6 vs. 5.9 Å, respectively. This suggests that erylosides B (226) bound more tightly to the Mpro complex than lopinavir.

Figure 6. Center-of-mass (CoM) distances (in Å) between erylosides B (black) and lopinavir (red) and GLN189 of Mpro over a 100 ns MD simulation.

2.3.4. Root-Mean-Square Deviation To watch the structural stability of the erylosides B (226)- and lopinavir-Mpro complexes, the root-mean-square deviation (RMSD) values of the backbone atoms of the whole complex were estimated (Figure 7). Unambiguously, the estimated RMSD values for the investigated complexes remained below 0.20 nm over the 100 ns MD simulations. The erylosides B (226)- and lopinavir-Mpro complexes reached the steady state in the first 10 ns MDs, and continued in an almost stationary state until the end of the simulations. Molecules 2021, 26, 2082 The current results emphasized that the erylosides B (226) is tightly bonded15 of 22 and does not impact the overall topology of SARS-CoV-2 Mpro.

Molecules 2021, 26, x FOR PEER REVIEW 15 of 22

FigureFigure 7. 7.Root-mean-square Root-mean-square deviation deviation (RMSD) (RMSD) of the backbone of the atomsbackbone from atoms the initial from structure the initial of structure of pro erylosideserylosides B (black)B (black) and and lopinavir lopinavir (red) with (red) M withthroughout Mpro throughout a 100 ns MD a simulation. 100 ns MD simulation. 2.4. Molecular Target Prediction and Network Analysis 2.4. Molecular Target Prediction and Network Analysis Erylosides B (226) protein targets associated with severe acute respiratory syndrome Erylosides B (226) protein targets associated with severe acute respiratory syndrome diseases were predicted using a SwissTargetPrediction DisGeNET online tool. One diseases were predicted using a SwissTargetPrediction DisGeNET online tool. One hundred hundred and seventeen genes were identified using Venn diagram comparison analysis. and seventeen genes were identiﬁed using Venn diagram comparison analysis. Commonly Commonlyshared genes shared for erylosides genes for Berylosides (226) were B BCL2L1,(226) were IL2, BCL2L1, PRKCA IL2,, and PRKCAPRKCB, and(Figure PRKCB8). (FigureSeveral 8). studies Several reported studies that reported these fourthat these genes four alter genes cytokine alter levels cytokine and levels immune and functions immune functionsin patients in with patients COVID-19 with COVID-19 infections infections and related and conditions related conditions [26,27]. Erylosides[26,27]. Erylosides B (226) Bpredicted (226) predicted gene targets gene were targets also analyzed were also via aanalyzed STRING via protein–protein a STRING interactionprotein–protein (PPI) interactionnetwork and (PPI) visualized network by and Cytoscape visualized 3.8.0. by TheCytoscape top 20 scored3.8.0. The genes top for 20 erylosides scored genes B (226) for erylosidesalso included B (226)BCL2L1, also included IL2, PRKCA BCL2L1,, and PRKCB IL2, PRKCA(Table, and S3). PRKCB (Table S3).

Figure 8. (A) Venn diagramdiagram analysisanalysis forfor erylosideserylosides B (226)(226) with SARS disease genes and (B)) STRINGSTRING PPIPPI networknetwork forfor the pro top targets identiﬁedidentified byby networknetwork analyzeranalyzer forfor erylosideserylosides BB (226)(226) asas aa MMpro inhibitor.inhibitor. 2.5. Pathway Enrichment Analysis (PEA) 2.5. Pathway Enrichment Analysis (PEA) Toward a deep dissection and mining of the erylosides B (226) target–function interac- Toward a deep dissection and mining of the erylosides B (226) target–function tions, PEA analysis and Boolean network modeling were conducted. A Voronoi treemap of interactions, PEA analysis and Boolean network modeling were conducted. A Voronoi erylosides B (226) top targeted pathway affected by the top 20 gene targets in response to treemap of erylosides B (226) top targeted pathway affected by the top 20 gene targets in erylosides B (226) in terms of SARS-CoV-2 infection was constructed (Figure9). Moreover, responsea Reactome to erylosides hierarchy B map (226) was in terms built toof provideSARS-CoV-2 a genome-wide infection was representation constructed (Figure of the 9).pathways Moreover, inﬂuenced a Reactome in response hierarchy to erylosides map B treatmentwas built (Figure to provide S4). Remarkably, a genome-wide among representation of the pathways influenced in response to erylosides B treatment (Figure S4). Remarkably, among the top 20 enriched pathways from the Reactome–PEA analyses and signal transduction, immune system and homeostasis signaling pathways were found to be the major pathways targeted by erylosides B (226), with a high significance (FDR < 0.00001%) (Table S4). Interestingly, under signal transduction, it was found that the G-protein coupled receptor (GPCR) signaling pathway was the most enriched pathway influenced by erylosides B (226) treatment within the human genome. Mining of the Reactome–PEA analysis results emphasized that a set of 15 genes (OPRK1, HTR2C, PRKCA, HTR2A, ADRA1D, PRKCB, DRD2, DRD3, ADRA2A, EDNRB, ADRA1A, ADRA2B, ADRA1B, ADRA2C, and F2) were significantly modulated as biological targets to erylosides B (226) as potent SARS-CoV-2 inhibitor. Additionally, the Reactome–PEA results revealed that these genes were found to interact with other genes/interactors, including P16220, P52292, P00533, P12931, P07550, P35414, P01019, P30559, P14416, and O15354.

Molecules 2021, 26, 2082 16 of 22

the top 20 enriched pathways from the Reactome–PEA analyses and signal transduction, Molecules 2021, 26, x FOR PEER REVIEW 16 of 22 immune system and homeostasis signaling pathways were found to be the major pathways targeted by erylosides B (226), with a high signiﬁcance (FDR < 0.00001%) (Table S4).

FigureFigure 9. 9.The The Voronoi Voronoi treemap treemap of of the the top top pathway pathway (signal (signal transduction) transduction) influenced influenced by theby topthe 20top gene 20 targetsgene targets in response in response to erylosides to erylosides B (226) B in(226) term in ofterm SARS-CoV-2 of SARS-CoV-2 infection. infection. The color The color highlights the highlights the over-representation of that pathway in the input dataset. Light grey signifies over-representation of that pathway in the input dataset. Light grey signifies pathways that are not pathways that are not significantly over-represented. significantly over-represented.

Interestingly,GPCRs are a undermassive signal family transduction, of surface ittransmembrane was found that receptors the G-protein in a coupledhuman cell re- ceptorcapable (GPCR) of responding signaling pathwayto numerous was the stimulants most enriched and pathwaypromoting influenced a cascade by erylosidesof cellular Bfunctions. (226) treatment According within to thethe humanGPCRs’ genome. pharmacolo Mininggical of attributes, the Reactome–PEA they are divided analysis into results four emphasizedprominent families: that a set the of rhodopsin-like 15 genes (OPRK1, receptors HTR2C, family, PRKCA, the secretin HTR2A, receptors ADRA1D, family, PRKCB, the DRD2,metabotropic DRD3, ADRA2A,glutamate/pherom EDNRB,one ADRA1A, receptors ADRA2B, family, ADRA1B,and the frizzled ADRA2C receptors, and F2 )family were significantly[28]. Remarkably, modulated the Reactome–PEA as biological targets analyses to erylosides results revealed B (226) asthat potent almost SARS-CoV-2 all GPCR inhibitor.prominent Additionally, families were the significantly Reactome–PEA influenced results revealedby erylosides that theseB (Figure genes 10). were Many found recent to interactstudies withfound other that genes/interactors, SARS-CoV-2 may including compromiseP16220, the P52292, GPCR P00533,signaling P12931, pathway P07550, and P35414,contribute P01019, to pulmonary P30559, P14416, edema’sand O15354pathophysi. ology [29]. Additionally, recent reports speculatedGPCRs that are a SARS-CoV-2 massive family might of surface hijack transmembrane the GPCR signaling receptors pathway— in a human mimicking cell capa- a bleprevious of responding case report to numerousin cholera toxin, stimulants ADP-ribosylation and promoting Gα as—to cascade activate of cellular CFTR and functions. switch Accordingon Cl- secretion, to the GPCRs’eventually pharmacological leading to dysregul attributes,ation they of lung are dividedion and intofluids four transmission, prominent families:which ultimately the rhodopsin-like led to lung receptors edema family,in COVID-19 the secretin patients receptors [30]. family, the metabotropic glutamate/pheromone receptors family, and the frizzled receptors family [28]. Remarkably, the Reactome–PEA analyses results revealed that almost all GPCR prominent families were significantly influenced by erylosides B (Figure 10). Many recent studies found that SARS-CoV-2 may compromise the GPCR signaling pathway and contribute to pulmonary edema’s pathophysiology [29]. Additionally, recent reports speculated that SARS-CoV-2 might hijack the GPCR signaling pathway— mimicking a previous case report in cholera toxin, ADP-ribosylation Gαs—to activate CFTR and switch on Cl- secretion, eventually leading to dysregulation of lung ion and fluids transmission, which ultimately led to lung edema in COVID-19 patients [30].

Figure 10. Graphic representation of the Reactome pathways influenced as a response to erylosides B (226) in term of SARS-CoV-2 infection. The representation showing the G-protein coupled receptor (GPCR) signaling pathway as the most enriched pathway influenced by erylosides B (226) treatment in the human genome.

Molecules 2021, 26, x FOR PEER REVIEW 16 of 22

Figure 9. The Voronoi treemap of the top pathway (signal transduction) influenced by the top 20 gene targets in response to erylosides B (226) in term of SARS-CoV-2 infection. The color highlights the over-representation of that pathway in the input dataset. Light grey signifies pathways that are not significantly over-represented.

GPCRs are a massive family of surface transmembrane receptors in a human cell ca- pable of responding to numerous stimulants and promoting a cascade of cellular functions. According to the GPCRs’ pharmacological attributes, they are divided into four prominent families: the rhodopsin-like receptors family, the secretin receptors family, the metabotropic glutamate/pheromone receptors family, and the frizzled receptors family [28]. Remarkably, the Reactome–PEA analyses results revealed that almost all GPCR prominent families were significantly influenced by erylosides B (Figure 10). Many recent studies found that SARS-CoV-2 may compromise the GPCR signaling pathway and contribute to pulmonary edema’s pathophysiology [29]. Additionally, recent reports speculated that SARS-CoV-2 might hijack the GPCR signaling pathway— mimicking a previous case report in cholera toxin, ADP-ribosylation Gαs—to activate CFTR and switch on Cl- Molecules 2021, 26, 2082 secretion, eventually leading to dysregulation of lung ion and fluids transmission, which17 of 22 ultimately led to lung edema in COVID-19 patients [30].

Figure 10. GraphicGraphic representation representation of of the the Reactome Reactome pathways pathways influenced inﬂuenced as as a a response response to to erylosides B B (226) in term of SARSSARS-CoV-2-CoV-2 infection. The representation showing the G G-protein-protein coupled receptor (GPCR) signaling pathway as the most enriched pathway influenced inﬂuenced by erylosides B (226) treatment in the human genome.

3. Materials and Methods 3.1. Mpro Preparation The resolved three-dimensional (3D) structure of SARS-CoV-2 main protease (Mpro) with a resolution of 2.16 Å (PDB code: 6LU7 [31]) complexed with peptidomimetic inhibitor (N3) was downloaded and used as a template for all molecular docking and MD calculations. The protein structure was prepared by removing all heteroatoms, crystallographic waters, and ions, conserving only the amino acid residues. The protonation states of amino acid residues of Mpro were assigned using H++ web server, and all missing hydrogen atoms were inserted [32]. The pKa values of Mpro amino acid residues were estimated under physical conditions of salinity = 0.15, internal dielectric constant = 10, pH = 7, and external dielectric constant = 80.

3.2. Inhibitor Preparation The chemical structures of the 227 scrutinized marine natural products (MNPs) de- scribed in literature as Red-Sea terpenes were obtained in SDF format from the PubChem database (https://pubchem.ncbi.nlm.nih.gov). All MNPs were processed using Omega2 software to generate the three-dimensional (3D) structures of the studied molecules [33,34]. Merck Molecular Force Field 94 (MMFF94S), implemented inside SZYBKI software [35,36], was applied to minimize the geometry of the investigated MNPs. The two-dimensional (2D) chemical structures of the scrutinized molecules are presented in Table S1.

3.3. Molecular Docking All molecular docking calculations were executed using AutoDock4.2.6 software [37]. The pdbqt ﬁle of SARS-CoV-2 main protease (Mpro) was prepared on the basis of the AutoDock protocol [38]. The maximum number of energy evaluations (eval) and the genetic-algorithm number (GA) were adjusted to 25,000,000 and 250, respectively. All other docking parameters were conserved at default. The grid box dimensions were adjusted to 60 Å × 60 Å × 60 Å to cover the SARS-CoV-2 Mpro binding pocket. Moreover, the grid spacing value was set to 0.375 Å. The coordinates of the grid center were located at −13.069, 9.740, and 68.490 in the x, y, and z directions, respectively. The atomic partial charges of the investigated MNPs were calculated using the Gasteiger method [39]. The predicted binding poses for each investigated MNP were processed by the built-in clustering analysis (1.0 Å RMSD tolerance), and the conformation with the lowest energy within the most massive cluster was picked out as a representative pose. Molecules 2021, 26, 2082 18 of 22

3.4. Molecular Dynamics Simulations AMBER16 software was employed to perform all MD simulations for the most potent MNPs complexed with SARS-CoV-2 Mpro [40]. Mpro and the investigated MNPs were determined using AMBER force field 14SB [41] and General AMBER force field (GAFF2) [42], respectively. In the present study, implicit-solvent and explicit-solvent MD simulations were performed. In the implicit-solvent MD simulations, the atomic partial charges of the investigated MNPs were evaluated using the AM1-BCC method [43]. No periodic boundary conditions and no cut-off were applied for nonbonded interactions (specifically, a cutoff value of 999 Å was employed). Furthermore, the solvent influence was estimated using igb = 1 solvent model [44]. Energy minimization was first carried out on the docked MNP-Mpro complexes for 500 steps, and the minimized complexes were then gently heated from 0 K to 300 K over 10 ps under NVT condition using a Langevin thermostat. Finally, the production stage was executed over 250 ps, and snapshots were assembled every 1 ps, giving 250 snapshots. In the present study, all implicit-solvent MD simulations were conducted using the CPU version of pmemd (pmemd.MPI) implemented inside AMBER16. In explicit-solvent MD simulations, the atomic partial charges of the investigated MNPs were estimated using the restrained electrostatic potential (RESP) approach at the HF/6-31G* level with the help of Gaussian09 software [45,46]. The TIP3P water model with periodic boundary conditions was utilized to solvate the MNP-Mpro complexes in a cubic water box with a minimum distance to the box edge of 15 Å. Energy minimizations for 5000 steps on the solvated MNP-Mpro complexes were performed using combined steepest and conjugate gradient methods. The minimized complexes were then smoothly heated from 0 to 300 K over 50 ps and a restraint of 10 kcal mol−1 Å−1 was applied on the main protease. The complexes were then equilibrated under NPT conditions for 1 ns. The production stage was then performed for each investigated Mpro-inhibitor complex over simulation times of 10 ns, 50 ns, and 100 ns. The particle mesh Ewald (PME) method was applied to estimate the long-range electrostatic forces and energies. Cut-off for coulombic interactions with a value of 12 Å was considered [47]. To maintain the temperature at 298 K, Langevin dynamics with a gamma_ln collision frequency set to 1.0 was utilized. A Berendsen barostat with a pressure relaxation time of 2 ps was applied for the pressure control [48]. To constrain all bonds including hydrogen atoms, a SHAKE algorithm with a time step of 2 fs was utilized [49]. For binding energy calculations and post-dynamics analyses, energy values and coordinates were collected every 10 ps over the production stage. All explicit-solvent MD simulations were executed utilizing the GPU version of pmemd (pmemd.cuda) implemented inside AMBER16. All in silico calculations involving quantum mechanics, molecular docking calculations, and MD simulations were carried out on the CompChem GPU/CPU cluster (hpc.compchem.net). BIOVIA DS Visualize 2020 was utilized to generate all molecular graphics [50].

3.5. Free Binding Energy Calculations The molecular mechanics/generalized Born surface area (MM/GBSA) approach [51] was applied to evaluate the binding free energies of the most potent MNPs complexed with SARS-CoV-2 Mpro. To appoint the polar solvation energy, the modiﬁed GB model proposed by Onufriev [52] (igb = 2) was employed. The uncorrelated snapshots collected throughout the MD simulations were subjected to MM/GBSA (∆Gbinding) energy calculations, which estimated as follows:

∆Gbinding = GComplex − (GMNP + GMpro) (1)

where the energy term (G) is evaluated as:

G = Eele + Evdw + GSA + GGB (2)

Eele and Evdw and are electrostatic and van der Waals energies, respectively. GSA is the nonpolar contribution to the solvation-free energy from the solvent-accessible surface Molecules 2021, 26, 2082 19 of 22

area (SASA). GGB is the electrostatic solvation free energy calculated from the generalized Born equation. A single-trajectory approach was applied, in which the coordinates of each MNP-Mpro,Mpro, and MNP were acquired from a single trajectory. Because of the expensive computational request and low prediction thoroughness in most cases, entropy calculations were ignored [53,54].

3.6. Protein–Protein Interaction and Pathway Enrichment Analysis (PEA) According to the structural resemblance of recognized inhibitor–target integrations, target forecasting of the most auspicious marine natural products was carried out utilizing the online web-based tools of SwissTargetPredicition (http://www.swisstargetprediction. ch). Moreover, the DisGeNET online database (https://www.disgenet.org) was applied to gather the accessible database for Severe Acute Respiratory Syndrome (SARS) diseases. The InteractiVenn online tool was employed to design Venn diagram [55]. A functional database of STRING for top predicted targets was applied to generate protein–protein interaction (PPI) network [56]. Cytoscape 3.8.2 was applied to examine target-function relation according to the network topological [57]. Furthermore, to explore all probable target–function relations for the top 20 SARS disease-related genes based on their network proﬁles, pathway enrichment analysis was achieved using Cytoscape 3.8.2 [57]. Lastly, the ReactomeFIViz tool was integrated for modeling and visualizing all potential drug–target interactions [58].

4. Conclusions In the context of the COVID-19 pandemic, it is apparent that zoonotic diseases pose a huge risk to public health and economic stability. A lack of targeted treatments has spurred the exploration of novel drug leads for which computational approaches provide a relatively quick and cost-effective approach. Herein, marine natural products isolated from the Red Sea were screened in silico as potential Mpro inhibitors. On the basis of molecular docking calculations, MD simulations, and molecular mechanics/generalized born surface area binding energy calculations, erylosides B (226) demonstrated a favorable pro binding afﬁnity with ∆Gbinding < −51.0 kcal/mol with M . The stability of 226 complexed with Mpro was conﬁrmed using energetic and structural analyses over the 100 ns MD simulation. Ultimately, we hypothesize that the utilization of erylosides B may represent a potential therapeutic approach against COVID-19 due to its useful capability to modulate the GPCR signaling pathway in COVID-19 patients. Experimental validation is expected to further identify the suitability of natural products such as erylosides B to serve as a SARS-CoV-2 inhibitor.

Supplementary Materials: The following are available online. Figure S1: 2D representations of interactions of lopinavir and the top 27 potent marine natural products (MNPs) with the proximal amino acid residues of SARS-CoV-2 main protease (Mpro), Figure S2: 3D representations of predicted binding modes of (i) 190, (ii) 178, (iii) 226 and (iv) lopinavir towards SARS-CoV-2 main protease (Mpro), Figure S3: 3D representations of binding modes of (i) erylosides B (226)- and (ii) lopinavir- Mpro complexes according to an average structure over a 100 ns MD simulation, Figure S4: a genome-wide Reactome hierarchy map of the pathways inﬂuenced by the top 20 gene targets in response to erylosides B (226) in term of SARS-CoV-2 infection, Table S1: evaluated docking score (in kcal/mol) for lopinavir and all investigated marine natural products (MNPs) against SARS-CoV-2 main protease (Mpro), Table S2: computed Autodock and MM/GBSA binding energies (in kcal/mol) for the top 27 potent marine natural products (MNPs) against SARS-CoV-2 main protease (Mpro) over 250 ps implicit solvent MD simulations, Table S3: network topological analysis for the predicted targets for erylosides B (226), Table S4: top 20 most relevant pathways for erylosides B (226) targets resulted from Pathway Enrichment Analysis (PEA). Author Contributions: Conceptualization, M.A.A.I. and M.-E.F.H.; Data curation, A.H.M.A. and K.A.A.A.; Formal analysis, A.H.M.A. and M.A.M.A.; Investigation, A.H.M.A. and M.-E.F.H.; Method- ology, M.A.A.I.; Project administration, M.A.A.I. and M.-E.F.H.; Resources, M.A.A.I.; Software, M.A.A.I.; Supervision, M.A.A.I.; Visualization, A.H.M.A. and M.A.M.A.; Writing—original draft, Molecules 2021, 26, 2082 20 of 22

A.H.M.A., K.A.A.A. and T.A.M.; Writing—review & editing, M.A.A.I., M.A.M.A., M.A.M.A.-H., E.M.E., M.F.M., F.A., K.S.A., H.R.E.-S., P.W.P., T.E. and M.-E.F.H. All authors have read and agreed to the published version of the manuscript. Funding: Mahmoud F. Moustafa extends his appreciation to the Deanship of Scientific Research at King Khalid University for funding this work under grant No. (R.G.P.1/143/42). H. R. El-Seedi is very grateful supported by the Swedish Research Council Vetenskapsrådet (VR-05885) and the Department of Molecular Biosciences, Wenner-Grens Institute, Stockholm University, Sweden, for the financial support. The computational work was completed with resources supported by the Science and Technology Development Fund, STDF, Egypt, Grants No. 5480 & 7972 (Granted to M.A.A.I). Mohamed Hegazy gratefully acknowledges the financial support from Alexander von Humboldt Foundation “Georg Foster Research Fellowship for Experienced Researchers”. Institutional Review Board Statement: Not applicable. Informed Consent Statement: Not applicable. Data Availability Statement: Data sharing is not applicable to this article. Conflicts of Interest: The authors declare no conflict of interest. Sample Availability: The samples are not available from the authors.

References 1. Vicenzi, E.; Canducci, F.; Pinna, D.; Mancini, N.; Carletti, S.; Lazzarin, A.; Bordignon, C.; Poli, G.; Clementi, M. Coronaviridae and SARS-associated coronavirus strain HSR1. Emerg. Infect. Dis. 2004, 10, 413–418. [CrossRef][PubMed] 2. Peng, Q.; Peng, R.; Yuan, B.; Zhao, J.; Wang, M.; Wang, X.; Wang, Q.; Sun, Y.; Fan, Z.; Qi, J.; et al. Structural and biochemical characterization of the nsp12-nsp7-nsp8 core polymerase complex from SARS-CoV-2. Cell Rep. 2020, 31, 107774–107788. [CrossRef] [PubMed] 3. Harapan, H.; Itoh, N.; Yufika, A.; Winardi, W.; Keam, S.; Te, H.; Megawati, D.; Hayati, Z.; Wagner, A.L.; Mudatsir, M. Coronavirus disease 2019 (COVID-19): A literature review. J. Infect. Public Health 2020, 13, 667–673. [CrossRef] 4. Huang, Y.; Yang, C.; Xu, X.F.; Xu, W.; Liu, S.W. Structural and functional properties of SARS-CoV-2 spike protein: Potential antivirus drug development for COVID-19. Acta Pharm. Sin. 2020, 41, 1141–1149. [CrossRef] 5. Dai, L.; Gao, G.F. Viral targets for vaccines against COVID-19. Nat. Rev. Immunol. 2021, 21, 73–82. [CrossRef] 6. Kemp, S.A.; Collier, D.A.; Datir, R.P.; Ferreira, I.; Gayed, S.; Jahun, A.; Hosmillo, M.; Rees-Spear, C.; Mlcochova, P.; Lumb, I.U.; et al. SARS-CoV-2 evolution during treatment of chronic infection. Nature 2021, 1–10. [CrossRef] 7. Zhang, L.; Lin, D.; Sun, X.; Curth, U.; Drosten, C.; Sauerhering, L.; Becker, S.; Rox, K.; Hilgenfeld, R. Crystal structure of SARS-CoV-2 main protease provides a basis for design of improved alpha-ketoamide inhibitors. Science 2020, 368, 409–412. [CrossRef][PubMed] 8. Kumar, Y.; Singh, H.; Patel, C.N. In silico prediction of potential inhibitors for the main protease of SARS-CoV-2 using molecular docking and dynamics simulation based drug-repurposing. J. Infect Public Health 2020, 13, 1210–1223. [CrossRef] 9. Ibrahim, M.A.A.; Abdelrahman, A.H.M.; Allemailem, K.S.; Almatroudi, A.; Moustafa, M.F.; Hegazy, M.-E.F. In silico evaluation of prospective anti-COVID-19 drug candidates as potential SARS-CoV-2 main protease inhibitors. Protein J. 2021, 1–14. [CrossRef] 10. Ibrahim, M.A.A.; Abdelrahman, A.H.M.; Hussien, T.A.; Badr, E.A.A.; Mohamed, T.A.; El-Seedi, H.R.; Pare, P.W.; Efferth, T.; Hegazy, M.F. In silico drug discovery of major metabolites from spices as SARS-CoV-2 main protease inhibitors. Comput. Biol. Med. 2020, 126, 104046–104055. [CrossRef] 11. Ibrahim, M.A.A.; Abdelrahman, A.H.M.; Hegazy, M.F. In-silico drug repurposing and molecular dynamics puzzled out potential SARS-CoV-2 main protease inhibitors. J. Biomol. Struct. Dyn. 2020, 1–12. [CrossRef][PubMed] 12. Ibrahim, M.A.A.; Abdeljawaad, K.A.A.; Abdelrahman, A.H.M.; Hegazy, M.F. Natural-like products as potential SARS-CoV-2 M(pro) inhibitors: In-silico drug discovery. J. Biomol. Struct. Dyn. 2020, 1–13. [CrossRef] 13. Ibrahim, M.A.A.; Mohamed, E.A.R.; Abdelrahman, A.H.M.; Allemailem, K.S.; Moustafa, M.F.; Shawky, A.M.; Mahzari, A.; Hakami, A.R.; Abdeljawaad, K.A.A.; Atia, M.A.M. Rutin and flavone analogs as prospective SARS-CoV-2 main protease inhibitors: In silico drug discovery study. J. Mol. Graph. Model. 2021, 107904. [CrossRef] 14. Zakaryan, H.; Arabyan, E.; Oo, A.; Zandi, K. Flavonoids: Promising natural compounds against viral infections. Arch. Virol. 2017, 162, 2539–2551. [CrossRef][PubMed] 15. Cherrak, S.A.; Merzouk, H.; Mokhtari-Soulimane, N. Potential bioactive glycosylated flavonoids as SARS-CoV-2 main protease inhibitors: A molecular docking and simulation studies. PLoS ONE 2020, 15, e0240653–e0240666. [CrossRef][PubMed] 16. Jo, S.; Kim, S.; Kim, D.Y.; Kim, M.-S.; Shin, D.H. Flavonoids with inhibitory activity against SARS-CoV-2 3CLpro. J. Enzym. Inhib. Med. Chem. 2020, 35, 1539–1544. [CrossRef] 17. Mayer, A.M.; Rodriguez, A.D.; Berlinck, R.G.; Hamann, M.T. Marine pharmacology in 2003–4: Marine compounds with anthelmintic antibacterial, anticoagulant, antifungal, anti-inflammatory, antimalarial, antiplatelet, antiprotozoal, antituberculosis, Molecules 2021, 26, 2082 21 of 22

and antiviral activities; affecting the cardiovascular, immune and nervous systems, and other miscellaneous mechanisms of action. Comp. Biochem. Physiol. C Toxicol. Pharm. 2007, 145, 553–581. 18. Mayer, A.M.; Hamann, M.T. Marine pharmacology in 2000: Marine compounds with antibacterial, anticoagulant, antifungal, anti-inflammatory, antimalarial, antiplatelet, antituberculosis, and antiviral activities; affecting the cardiovascular, immune, and nervous systems and other miscellaneous mechanisms of action. Mar. Biotechnol. 2004, 6, 37–52. 19. Hegazy, M.E.; Mohamed, T.A.; Alhammady, M.A.; Shaheen, A.M.; Reda, E.H.; Elshamy, A.I.; Aziz, M.; Pare, P.W. Molecular architecture and biomedical leads of terpenes from red sea marine invertebrates. Mar. Drugs 2015, 13, 3154–3181. [CrossRef] 20. Koop, K.; Booth, D.; Broadbent, A.; Brodie, J.; Bucher, D.; Capone, D.; Coll, J.; Dennison, W.; Erdmann, M.; Harrison, P.; et al. ENCORE: The effect of nutrient enrichment on coral reefs. Synthesis of results and conclusions. Mar. Pollut. Bull. 2001, 42, 91–120. [CrossRef] 21. Ayyad, S.N.; Alarif, W.M.; Al-Footy, K.O.; Selim, E.A.; Ghandourah, M.A.; Aly, M.M.; Alorfi, H.S. Isolation, antimicrobial and antitumor activities of a new polyhydroxysteroid and a new diterpenoid from the soft coral Xenia umbellata. Z Nat. C J. Biosci. 2017, 72, 27–34. [CrossRef][PubMed] 22. Hegazy, M.E.; Gamal Eldeen, A.M.; Shahat, A.A.; Abdel-Latif, F.F.; Mohamed, T.A.; Whittlesey, B.R.; Pare, P.W. Bioactive hydroperoxyl cembranoids from the Red Sea soft coral Sarcophyton glaucum. Mar. Drugs 2012, 10, 209–222. [CrossRef] 23. Horby, P.W.; Mafham, M.; Bell, J.L.; Linsell, L.; Staplin, N.; Emberson, J.; Palfreeman, A.; Raw, J.; Elmahi, E.; Prudon, B.; et al. Lopinavir–ritonavir in patients admitted to hospital with COVID-19 (RECOVERY): A randomised, controlled, open-label, platform trial. Lancet 2020, 396, 1345–1352. [CrossRef] 24. De Vivo, M.; Masetti, M.; Bottegoni, G.; Cavalli, A. Role of molecular dynamics and related methods in drug discovery. J. Med. Chem. 2016, 59, 4035–4061. [CrossRef] 25. Kerrigan, J.E. Molecular dynamics simulations in drug design. In In Silico Models for Drug Discovery; Kortagere, S., Ed.; Humana Press: Totowa, NJ, USA, 2013; pp. 95–113. [CrossRef] 26. Noroozi, R.; Branicki, W.; Pyrc, K.; Labaj, P.P.; Pospiech, E.; Taheri, M.; Ghafouri-Fard, S. Altered cytokine levels and immune responses in patients with SARS-CoV-2 infection and related conditions. Cytokine 2020, 133, 155143–155147. [CrossRef] 27. Zhou, Y.; Hou, Y.; Shen, J.; Huang, Y.; Martin, W.; Cheng, F. Network-based drug repurposing for novel coronavirus 2019- nCoV/SARS-CoV-2. Cell Discov. 2020, 6, 14. [CrossRef][PubMed] 28. Oldham, W.M.; Hamm, H.E. Heterotrimeric G protein activation by G-protein-coupled receptors. Nat. Rev. Mol. Cell Biol. 2008, 9, 60–71. [CrossRef] 29. Birch, C.A.; Molinar-Inglis, O.; Trejo, J. Subcellular hot spots of GPCR signaling promote vascular inflammation. Curr. Opin. Endocr. Metab. Res. 2021, 16, 37–42. [CrossRef] 30. Hameid, R.A.; Cormet-Boyaka, E.; Kuebler, W.M.; Uddin, M.; Berdiev, B.K. SARS-CoV-2 may hijack GPCR signaling pathways to dysregulate lung ion and fluid transport. Am. J. Physiol. Lung Cell Mol. Physiol. 2021, 320, L430–L435. [CrossRef] 31. Jin, Z.; Du, X.; Xu, Y.; Deng, Y.; Liu, M.; Zhao, Y.; Zhang, B.; Li, X.; Zhang, L.; Peng, C.; et al. Structure of M(pro) from SARS-CoV-2 and discovery of its inhibitors. Nature 2020, 582, 289–293. [CrossRef][PubMed] 32. Gordon, J.C.; Myers, J.B.; Folta, T.; Shoja, V.; Heath, L.S.; Onufriev, A. H++: A server for estimating pKas and adding missing hydrogens to macromolecules. Nucleic Acids Res. 2005, 33, W368–W371. [CrossRef] 33. Hawkins, P.C.; Skillman, A.G.; Warren, G.L.; Ellingson, B.A.; Stahl, M.T. Conformer generation with OMEGA: Algorithm and validation using high quality structures from the Protein Databank and Cambridge Structural Database. J. Chem. Inf. Model. 2010, 50, 572–584. [CrossRef][PubMed] 34. OMEGA 2.5.1.4; OpenEye Scientific Software: Santa Fe, NM, USA, 2013. 35. SZYBKI 1.9.0.3; OpenEye Scientific Software: Santa Fe, NM, USA, 2016. 36. Halgren, T.A. MMFF VI. MMFF94s option for energy minimization studies. J. Comput. Chem. 1999, 20, 720–729. [CrossRef] 37. Morris, G.M.; Huey, R.; Lindstrom, W.; Sanner, M.F.; Belew, R.K.; Goodsell, D.S.; Olson, A.J. AutoDock4 and AutoDockTools4: Automated docking with selective receptor flexibility. J. Comput. Chem. 2009, 30, 2785–2791. [CrossRef][PubMed] 38. Forli, S.; Huey, R.; Pique, M.E.; Sanner, M.F.; Goodsell, D.S.; Olson, A.J. Computational protein-ligand docking and virtual drug screening with the AutoDock suite. Nat. Protoc. 2016, 11, 905–919. [CrossRef] 39. Gasteiger, J.; Marsili, M. Iterative partial equalization of orbital electronegativity—A rapid access to atomic charges. Tetrahedron 1980, 36, 3219–3228. [CrossRef] 40. Case, D.A.; Betz, R.M.; Cerutti, D.S.; Cheatham, T.E.; Darden, T.A.; Duke, R.E.; Giese, T.J.; Gohlke, H.; Goetz, A.W.; Homeyer, N.; et al. AMBER 2016; University of California: San Francisco, CA, USA, 2016. 41. Maier, J.A.; Martinez, C.; Kasavajhala, K.; Wickstrom, L.; Hauser, K.E.; Simmerling, C. ff14SB: Improving the accuracy of protein side chain and backbone parameters from ff99SB. J. Chem. Theory Comput. 2015, 11, 3696–3713. [CrossRef] 42. Wang, J.; Wolf, R.M.; Caldwell, J.W.; Kollman, P.A.; Case, D.A. Development and testing of a general amber force field. J. Comput. Chem. 2004, 25, 1157–1174. [CrossRef][PubMed] 43. Jakalian, A.; Jack, D.B.; Bayly, C.I. Fast, efficient generation of high-quality atomic charges. AM1-BCC model: II. Parameterization and validation. J. Comput. Chem. 2002, 23, 1623–1641. [CrossRef] 44. Roux, B.; Simonson, T. Implicit solvent models. Biophys Chem. 1999, 78, 1–20. [CrossRef] 45. Bayly, C.I.; Cieplak, P.; Cornell, W.D.; Kollman, P.A. A well-behaved electrostatic potential based method using charge restraints for deriving atomic charges—The RESP model. J. Phys. Chem. 1993, 97, 10269–10280. [CrossRef] Molecules 2021, 26, 2082 22 of 22

46. Frisch, M.J.; Trucks, G.W.; Schlegel, H.B.; Scuseria, G.E.; Robb, M.A.; Cheeseman, J.R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G.A.; et al. Gaussian 09; Revision E01; Gaussian Inc.: Wallingford, CT, USA, 2009. 47. Darden, T.; York, D.; Pedersen, L. Particle mesh Ewald: AnN·log(N) method for Ewald sums in large systems. J. Chem. Phys. 1993, 98, 10089–10092. [CrossRef] 48. Berendsen, H.J.C.; Postma, J.P.M.; Vangunsteren, W.F.; Dinola, A.; Haak, J.R. Molecular-dynamics with coupling to an external bath. J. Chem Phys. 1984, 81, 3684–3690. [CrossRef] 49. Miyamoto, S.; Kollman, P.A. Settle—An Analytical Version of the Shake and Rattle Algorithm for Rigid Water Models. J. Comput. Chem. 1992, 13, 952–962. [CrossRef] 50. Dassault Systèmes BIOVIA, B.D.S.V., Version 2019; Dassault Systèmes BIOVIA: San Diego, CA, USA, 2019. 51. Massova, I.; Kollman, P.A. Combined molecular mechanical and continuum solvent approach (MM-PBSA/GBSA) to predict ligand binding. Perspect Drug Discov. Des. 2000, 18, 113–135. [CrossRef] 52. Onufriev, A.; Bashford, D.; Case, D.A. Exploring protein native states and large-scale conformational changes with a modiﬁed generalized born model. Proteins 2004, 55, 383–394. [CrossRef] 53. Hou, T.; Wang, J.; Li, Y.; Wang, W. Assessing the performance of the molecular mechanics/Poisson Boltzmann surface area and molecular mechanics/generalized Born surface area methods. II. The accuracy of ranking poses generated from docking. J. Comput. Chem. 2011, 32, 866–877. [CrossRef][PubMed] 54. Wang, E.; Sun, H.; Wang, J.; Wang, Z.; Liu, H.; Zhang, J.Z.H.; Hou, T. End-Point Binding Free Energy Calculation with MM/PBSA and MM/GBSA: Strategies and Applications in Drug Design. Chem. Rev. 2019, 119, 9478–9508. [CrossRef][PubMed] 55. Heberle, H.; Meirelles, G.V.; da Silva, F.R.; Telles, G.P.; Minghim, R. InteractiVenn: A web-based tool for the analysis of sets through Venn diagrams. BMC Bioinform. 2015, 16, 169–175. [CrossRef][PubMed] 56. Li, R.; Ma, X.; Song, Y.; Zhang, Y.; Xiong, W.; Li, L.; Zhou, L. Anti-colorectal cancer targets of resveratrol and biological molecular mechanism: Analyses of network pharmacology, human and experimental data. J. Cell Biochem. 2019, 120, 11265–11273. [CrossRef] 57. Shannon, P.; Markiel, A.; Ozier, O.; Baliga, N.S.; Wang, J.T.; Ramage, D.; Amin, N.; Schwikowski, B.; Ideker, T. Cytoscape: A software environment for integrated models of biomolecular interaction networks. Genome Res. 2003, 13, 2498–2504. [CrossRef] [PubMed] 58. Blucher, A.S.; McWeeney, S.K.; Stein, L.; Wu, G. Visualization of drug target interactions in the contexts of pathways and networks with ReactomeFIViz. F1000Research 2019, 8, 908. [CrossRef][PubMed]