key: cord-0964450-yptikz5q authors: Zaki, Ahmed A.; Ashour, Ahmed; Elhady, Sameh S.; Darwish, Khaled M.; Al-Karmalawy, Ahmed A. title: Calendulaglycoside A Showing Potential Activity Against SARS-CoV-2 Main Protease: Molecular Docking, Molecular Dynamics, and SAR Studies date: 2021-05-17 journal: J Tradit Complement Med DOI: 10.1016/j.jtcme.2021.05.001 sha: b32513d62d35357a784ddc0b56016e7e0e385650 doc_id: 964450 cord_uid: yptikz5q BACKGROUND AND AIM: The discovery of drugs capable of inhibiting SARS-CoV-2 is a priority for human beings due to the severity of the global health pandemic caused by COVID-19. To this end, natural products can provide therapeutic alternatives that could be employed as an effective safe treatment for COVID-19. EXPERIMENTAL PROCEDURE: Twelve compounds were isolated from the aerial parts of C. officinalis L. and investigated for their inhibitory activities against SARS-CoV-2 M(pro) compared to its co-crystallized N3 inhibitor using molecular docking studies. Furthermore, a 100 ns MD simulation was performed for the most active two promising compounds, Calendulaglycoside A (SAP5) and Osteosaponin-I (SAP8). RESULTS AND CONCLUSION: At first, molecular docking studies showed interesting binding scores as compared to the N3 inhibitor. Calendulaglycoside A (SAP5) achieved a superior binding than the co-crystallized inhibitor indicating promising affinity and intrinsic activity towards the M(pro) of SARS-CoV-2 as well. Moreover, findings illustrated preferential stability for SAP5 within the M(pro) pocket over that of N3 beyond the 40 ns MD simulation course. Structural preferentiality for triterpene-M(pro) binding highlights the significant role of 17β-glucosyl and carboxylic 3α-galactosyl I moieties through high electrostatic interactions across the MD simulation trajectories. Furthermore, this study clarified a promising SAR responsible for the antiviral activity against the SARS-CoV-2 M(pro) and the design of new drug candidates targeting it as well. The above findings could be promising for fast examining the previously isolated triterpenes both pre-clinically and clinically for the treatment of COVID-19. The existence of the COVID-19 pandemic in the world demands the need to identify and 30 characterize new drug candidates to address the health problem caused by the SARS-CoV-2. 1, 2 31 After SARS-CoV-2 infection, two overlapping polyproteins are produced (replicase 1a and 1ab) 32 which create many functional subunits for viral replication. This process requires two internally 33 encoded proteases that cleave the aforementioned polyproteins at specific regions. The main ulcers. 18 In the middle ages, Calendula flowers were used for liver obstructions, snake bites and 66 to strengthen the heart. It was used in the 18 th century as a remedy for headaches, jaundice, and 67 red eyes. The plant was employed in the civil war to treat wounds and as a remedy for measles, 68 smallpox, and jaundice. 13 69 Furthermore, sesquiterpene glycosides of Calendula were documented to inhibit the 70 replication of the Herpes virus and rhinovirus. 19, 20 Also, Calendula extract exhibited antiviral 71 activity against influenza and herpes simplex viruses. 21 Nowadays, computational drug design methods such as molecular docking and molecular 73 dynamics are very promising tools in the discovery of new drug candidates. 22 Also, 74 bioinformatics techniques opened new horizons to detect the crucial key amino acids at nearly 75 identical physiological conditions, and so confirm greatly the results of computational methods. 76 However, new drug candidates can be introduced depending on the chemical nature of the drug 77 and its target receptor, saving effort and cost as well. 23 78 Therefore, as an extension to our previously promising work [24] [25] [26] [27] [28] [29] , and depending on the 105 The air-dried and powdered aerial parts of C. officinalis L. (500 g) were extracted till 106 exhaustion with methanol at room temperature. The collected solvent was evaporated to dryness 107 using a rotary evaporator. The combined crude extract (total, 67 g) was suspended in 150 mL 108 water and defatted with hexanes (0.5 L x 4) to yield the hex-fraction (7.8 gm) and 57 gm 109 representing the remaining extract. The silica gel column of hex-fraction using the gradients of 5 110 -25 % ethyl acetate in petroleum ether resulted in the isolation of compound 11 (17 mg). 111 The protein data bank was used to obtain the crystal structure of the SARS-CoV-2 main 142 protease (M pro ). The selected enzyme was downloaded containing its co-crystallized (N3) 143 inhibitor (PDB code 6LU7) 30 . It was protonated where hydrogen atoms were added with their 3D 144 geometry, corrected for any found errors in the connection or type of different atoms, and then 145 energy minimized at the end of the preparation steps. Moreover, Site Finder was applied to 146 define and isolate the same binding pocket of the co-crystallized native inhibitor (N3) as dummy 147 atoms for the docking step. 35 149 At first, a validation process to ensure the accuracy of our docking program was run, and 150 the valid behavior of the co-crystallized (N3) inhibitor was confirmed by a low RMSD value of 151 1.23 Å as depicted in the supplementary data ( Figure SI2) . 37, 38 The following methodology 152 was performed: the previously prepared active site was loaded, and the docking process was 153 started as a general one. The docking site was adjusted as dummy atoms, the forcefield was 154 selected to be Amber10:EHT, the rigid receptor was selected as refinement methodology and 155 GBVI/WSA dG as its scoring methodology, the placement methodology was as triangle matcher 156 and its scoring methodology was applied as London dG. The prepared MDB file containing the 157 twelve prepared medicinal plant ligands (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) together with the co-crystallized inhibitor (N3, 158 13) was loaded and then the general dock calculations were run automatically 39 to select poses 159 with the best scores, rmsd_refine values, and binding modes. 161 The top-dock models of the most promising leads (Calendulaglycoside A (SAP5) and 162 Osteosaponin-I (SAP8)), as well as N3 in complex with SARS-CoV-2 main protease, were 163 chosen as starting coordinates for 100 ns all-atoms molecular dynamic (MD) simulation using a 164 GROMACS-2019 software package (GNU, General Public License; http://www.gromacs.org) 165 and CHARMM36 force field 32 . 28, 29 Regarding the investigated ligands, the CHARMM force 166 field parameters were automatically generated using the CHARMM General Force Field 167 (CGenFF) program (ParamChem project; https://cgenff.umaryland.edu/ ). 40 Each ligand-protein 168 complex was then solvated within a cubic box of the transferable intermolecular potential with a 169 three-points (TIP3P) water model (100 x 100 x 100 Å) allowing a minimum of 10 Å marginal 170 distance between protein and each side of the 3D box. 41 Under periodic boundary conditions 171 J o u r n a l P r e -p r o o f implementation, the protein residues were assigned for their standard ionization states at 172 physiological conditions (pH 7.0), and the whole complexes were neutralized via sufficient 173 numbers of K‫‬ + and Clions added via Monte-Carlo ion-placing method. 42 The total number of 174 atoms, including water and ions, for each final system, was 89112, 85645, and 85577 atoms for 175 N3, SAP5, and SAP8 systems, respectively. The MD simulation was conducted over three stages 176 (minimization, equilibration, and production) and a 1000 KJ/mol.nm 2 force constant was used 177 along minimization and equilibration stages for restraining all heavy atoms and preserving 178 original protein folding. 43 The first stage involved initial optimization of each system geometry 179 using 5,000 iterations (5 ps) with the steepest descent algorithm. The subsequent step involved Waals interactions were truncated at 10 Å using the Verlet cut-off scheme. 48 Computing comparative data, including root-mean-square deviation (RMSD), difference Both the pharmacokinetic and pharmacodynamic properties which play an essential role 224 in the optimization of lead molecules were studied for the tested isolated molecules using the 225 Swiss ADME server. Prediction of ADME (Absorption, Distribution, Metabolism, and 233 Using the compounds; Machaerinic acid 3-O-β-D-glucuronopyranoside (1) 53 , 234 Calendulaglycoside C (2) 54 , Glycoside F (3) 55 , Calenduloside G (4) 56 , Calendulaglycoside A 235 (5) 57 , Calenduloside B (6) 54 , Glucoside I (7) 58 , Osteosaponin-I (8) 59 , Arvensoside B (9), 236 Oleanolic acid (10) 58 and Stigmasterol (11), which are previously isolated by Zaki and Qiu 53 as 237 standard metabolites in the chromatographic re-investigation of total extract of C. officinalis, 238 compound 12 was isolated. It was obtained as a white powder with the molecular formula of The SARS-CoV-2‫‬main protease catalytic dyad is composed of Cys-His, and the binding 247 site is located between subunits I and II. 61 Molecular docking of the compounds (1-12) and N3 248 inhibitor 13 into M pro active site was done. They got stabilized at the inhibitor binding site by 249 diverse interactions ( Table 1) . The obtained molecular docking results were considered valid 250 since redocking the crystalized ligand (N3) furnished great ligand superposition with root-mean 251 standard deviation (RMSD) below 2 Å. 62 The descending strength order based on the score 252 values: (5) ˃ N3 inhibitor (13, redocked) ˃ (8) ˃ (4) ˃ (2) ˃ (6) ˃ (3) ˃ (1) ˃ (9) ˃ (12) ˃ (7) ˃ 253 (10) ˃ (11). Surprisingly, SAP5 was more promising with a binding score of (-9.90 Kcal/mole) 254 compared to the N3 inhibitor (-9.37 Kcal/mole). Studying the obtained docking results of the tested triterpenes, it was obvious that many compounds (1-12) and the redocked N3 (13) were described in detail in Table 1 . Furthermore, 273 the 2D and 3D binding interactions, surface and maps, and 3D protein pocket positioning for all 274 the tested compounds are represented in the Supplementary data (Figures SI3 and SI4) . 275 However, the 3D binding interactions and the 3D pocket positioning of the best selected two 276 isolated triterpene compounds (SAP5 and SAP8) compared to the redocked N3 inhibitor (13) are 277 represented in Table 2 . 278 interactions with important residues. 63 In these regards, both SAP5 and SAP8, in complex with 288 SARS-CoV-2 M pro , were enrolled within 100 ns all-atom MD simulation runs. to the other two ligands (Figure 3 ). This may be rationalized for the poor ligand-protein 417 interaction and ligand accommodation at the S1' and S3 sub-pockets across the MD simulation 418 run. Extensive analysis of the obtained ΔRMSF values permits comparative analysis of the 420 active site residue mobility across the three complexed proteins. Interestingly, only the catalytic 421 His41, of the S1' sub-pocket residues, exhibited decreased mobility across the three proteins, 422 with the highest ΔRMSF value was for all N3-bound proteins ( Table 3) . However, the other S1' In the presented study, findings from the Rg analysis appear to ensure the preferential 461 stability of the SAP5-protein complex previously presented by the RMSD trajectory analysis. The protein Rg values were, to some extent, comparable among the investigated systems being 463 fluctuated around close averages ( Figure 4A and Table 4 ). Nevertheless, the ligand Rg traces and N3-bounded proteins, respectively. The latter demonstrates the stability of solvent-exposed Being solvent-exposed, several polar interactions with surface hydrophilic residues, Asn72 and 638 Arg76, permitted the stabilization of the N3 at its new site. Moving towards the other investigated triterpene, SAP8 was bounded at proximity to the 640 binding site depicted by N3 at the end of the MD simulation run (Figure 7C) . Nevertheless, the 641 transition of SAP8, from the initial docked orientation to its final state, was proceeded more Thus, the MD-directed MM/PBSA approach was performed for binding-free energy calculations. Since free energy calculation is an equilibrium thermodynamic metric, examining the plots of 697 energy (total, potential and kinetic) and temperature versus time for all systems were also 698 beneficial to further ensure system stability across the designated 100-ns time frames. The time 699 evolution for system energies and temperature was steady across the MD simulation times within 700 -1.18x10 6 to -1.16 x10 6 kJ/mol, 2.17x10 5 to 2.24x10 5 kJ/mol, -9.60x10 5 to -9.34 x10 5 kJ/mol, 701 and 298 to 307 K for potential, kinetic and total energies as well as temperature (Figure 8 ). System stability was significant for all three systems, with slightly lower negative values for N3 703 total energy term and comparable values for both SAP5 and SAP8 systems. For further evaluation of the comparative ligand-residues interactions at M pro active binding 786 site for SAP5 and N3, the binding-energy decomposition was used to identify key residues 787 involved within obtained ΔE binding (Figure 8) . 51 At 10-40 ns time frame, several residues, which 788 have participated within the initially docked complexes, showed significant contributions in the 789 calculated binding-free energy. For the SAP5-protein complex, the highest residue-binding 790 energy contribution was for S1 Glu166 confirming the key role of this residue in small molecule 791 binding to the SARS-CoV-2 M pro target ( Figure 8A) showed the greatest contribution within the ΔE binding for the N3-protein complex ( Figure 8B ). Second in order, both Glu166, at S1 sub-pocket, Gln192. Comparing the above 10-20 ns residue-wise contributions with that at 30-60 ns, showed 817 that several pocket residues exhibited compromised energy contributions. These residues include 818 the S1' catalytic Cys145, S2 Met165, S1 Glu166, and S2 Arg188. Compromised Glu166 energy 819 contribution in case of N3, however, is maintained with SAP5 system, highlights the superior 820 stability of SAP5 since several studies report Glu166 being indispensable for M pro binding. (Figures SI6 and SI7) . We wish to confirm that there are no known conflicts of interest associated with this publication and there has been no significant financial support for this work that could have influenced its outcome. We confirm that the manuscript has been read and approved by the five authors and that there are no other persons who satisfied the criteria for authorship but are not listed. We further confirm that the order of authors listed in the manuscript has been approved by all of us. We confirm that we have given due consideration to the protection of intellectual property associated with this work and that there are no impediments to publication, including the timing of publication, concerning intellectual property. In so doing we confirm that we have followed the regulations of our institutions concerning intellectual property. We understand that the Corresponding Author is the sole contact for the Editorial process. He is responsible for communicating with the other authors about progress, submissions of revisions and final approval of proofs. We confirm that we have provided current, correct email address which is accessible by the Corresponding Author and which has been configured to accept emails from akarmalawy@horus.edu.eg Progress in developing inhibitors of SARS-CoV-2 3C-like protease Molecular docking and dynamics simulation revealed the potential inhibitory activity of ACEIs 905 against SARS-CoV-2 targeting hACE2 receptor Journal of chemical 907 information and modeling 2020 CoV natural products with the potential to inhibit SARS-CoV-2 (COVID-19) Structure and in vitro antiviral activity 959 of sesquiterpene glycosides from Calendula arvensis Study of 961 antiviral properties of Calendula officinalis Computational approaches for drug discovery. Multidisciplinary Digital Publishing 963 Institute: 2019 A review on applications of computational methods in drug screening and 965 design Molecular docking reveals the 967 potential of Cleome amblyocarpa isolated compounds to inhibit COVID-19 virus main protease Molecular docking, molecular 970 dynamics, and in vitro studies reveal the potential of angiotensin II receptor blockers to inhibit the 971 COVID-19 main protease Molecular docking and 973 dynamics simulations reveal the potential of anti-HCV drugs to inhibit COVID-19 main protease Revisiting activity of some glucocorticoids as a potential inhibitor of SARS-CoV-2 main protease: 977 theoretical study Molecular Docking and Dynamics Simulation Revealed the Potential Inhibitory Activity of 980 ACEIs Against SARS-CoV-2 Targeting the hACE2 Receptor In a search for potential drug candidates for combating COVID-19: 984 computational study revealed salvianolic acid B as a potential therapeutic targeting 3CLpro and spike 985 proteins Structure 988 of M pro from SARS-CoV-2 and discovery of its inhibitors Development of parameters compatible with the CHARMM36 force field for [Fe4S4] 2+ clusters and 993 molecular dynamics simulations of adenosine-5'-phosphosulfate reductase in GROMACS 2019 Molecular modelling of mebendazole polymorphs as a 996 potential colchicine binding site inhibitor Revisiting Activity of Some Nocodazole Analogues as a 998 Potential Anticancer Drugs Using Molecular Docking and DFT Calculations Tanshinone IIA synergistically enhances the antitumor activity of doxorubicin by interfering with the 1002 PI3K/AKT/mTOR pathway and inhibition of topoisomerase II: in vitro and molecular docking studies Bioassay-guided isolation of a new cytotoxic ceramide from Cyperus rotundus L RosettaLigand docking with full ligand and receptor flexibility The performance of current methods in ligand-1010 protein docking Empagliflozin and Doxorubicin 1012 Synergistically Inhibit the Survival of Triple-Negative Breast Cancer Cells via Interfering with the mTOR 1013 Pathway and Inhibition of Calmodulin: In Vitro and Molecular Docking Studies CHARMM general force field: A force field for drug-like molecules 1017 compatible with the CHARMM all-atom additive biological force fields Building water models: a different approach. The 1020 journal of physical chemistry letters Biomolecular simulations 1022 under realistic macroscopic salt conditions Molecular basis of the potential interaction of SARS-CoV-2 spike protein to CD147 in 1026 COVID-19 associated-lymphopenia Dynamic attractor for the Berendsen thermostat an the slow dynamics of 1028 biomacromolecules An approach to developing a force field for molecular 1030 simulation of martensitic phase transitions between phases with subtle differences in energy and 1031 structure Particle mesh Ewald: An N⋅ log (N) method for Ewald sums in 1033 large systems LINCS: a linear constraint solver for 1035 molecular simulations A flexible algorithm for calculating pair interactions on SIMD architectures Journal of molecular 1039 graphics The PyMOL molecular graphics system, version 1.8 g_mmpbsa A GROMACS tool for 1042 high-throughput MM-PBSA calculations SwissADME: a free web tool to evaluate pharmacokinetics drug-likeness and medicinal chemistry friendliness of small molecules Machaerinic acid 3-O-β-D-glucuronopyranoside from Calendula officinalis Plant metabolites Bacoside A3 A triterpenoid saponin from Bacopa 1052 monniera Structures of calendulosides G and H from 1054 the roots of Calendula officinalis Anti-inflammatory, anti-1056 tumor-promoting, and cytotoxic activities of constituents of marigold (Calendula officinalis) flowers. 1057 Journal of natural products Antibacterial and antiparasitic activity of oleanolic acid and its glycosides isolated from marigold 1060 (Calendula officinalis) saponin glycosides from Osteospermum vaillantii Oleanolic acid 3-O-beta-D-glucuronopyranoside from 1064 the roots of Calendula officinalis Crystal structure of SARS-CoV-2 main protease provides a basis for design of improved α-1067 ketoamide inhibitors Evaluation of docking performance: comparative 1069 data on docking algorithms Molecular dynamics simulations in biology Relaxation estimation of RMSD in molecular 1073 dynamics immunosimulations. Computational and mathematical methods in medicine 2012 Structure Of Biomolecules Through Molecular 1076 Principal component analysis: a method for determining the essential 1078 dynamics of proteins Molecular dynamics simulations and principal component 1080 analysis on human laforin mutation W32G and W32G/K87A. The protein journal Structural insights into conformational stability of both wild-1082 type and mutant EZH2 receptor Alanine 1084 mutation of the catalytic sites of Pantothenate Synthetase causes distinct conformational changes in the 1085 ATP binding region Structural dynamics flexibility informs function and 1087 evolution at a proteome scale 3-Acyltetramic acids as a novel class of inhibitors for human 1090 kallikreins 5 and 7 Structural flexibility and protein adaptation 1092 to temperature: Molecular dynamics analysis of malate dehydrogenases of marine molluscs Insights from 1095 molecular dynamics simulations: structural basis for the V567D mutation-induced instability of zebrafish 1096 alpha-dystroglycan and comparison with the murine model A statistical approach to the interpretation 1098 of molecular dynamics simulations of calmodulin equilibrium dynamics Binding free energy calculation using quantum mechanics aimed for drug lead 1101 optimization Data driven inference for the repulsive exponent of the Lennard-Jones potential in 1105 molecular dynamics simulations Molecular dynamics simulations of Coulomb explosion Methods in Physics Research Section B: Beam Interactions with Materials and Atoms the introduction of a glucose moiety at the C-28 carboxylic acid group enhances the main 881 protease binding affinity as well. 894 The authors declare that there is no conflict of interest.