Docking Analysis of Some Bioactive Compounds from Traditional Plants against SARS-CoV-2 Target Proteins

COVID-19 is still a global pandemic that has not been stopped. Many traditional medicines have been demonstrated to be incredibly helpful for treating COVID-19 patients while fighting the disease worldwide. We introduced 10 bioactive compounds derived from traditional medicinal plants and assessed their potential for inhibiting viral spike protein (S-protein), Papain-like protease (PLpro), and RNA dependent RNA polymerase (RdRp) using molecular docking protocols where we simulate the inhibitors bound to target proteins in various poses and at different known binding sites using Autodock version 4.0 and Chimera 1.8.1 software. Results found that the chicoric acid, quinine, and withaferin A ligand strongly inhibited CoV-2 S -protein with a binding energy of −8.63, −7.85, and −7.85 kcal/mol, respectively. Our modeling work also suggested that curcumin, quinine, and demothoxycurcumin exhibited high binding affinity toward RdRp with a binding energy of −7.80, −7.80, and −7.64 kcal/mol, respectively. The other ligands, namely chicoric acid, demothoxycurcumin, and curcumin express high binding energy than the other tested ligands docked to PLpro with −7.62, −6.81, and −6.70 kcal/mol, respectively. Prediction of drug-likeness properties revealed that all tested ligands have no violations to Lipinski’s Rule of Five except cepharanthine, chicoric acid, and theaflavin. Regarding the pharmacokinetic behavior, all ligand predicted to have high GI-absorption except chicoric acid and theaflavin. At the same way chicoric acid, withaferin A, and withanolide D predicted to be substrate for multidrug resistance protein (P-gp substrate). Caffeic acid, cepharanthine, chicoric acid, withaferin A, and withanolide D also have no inhibitory effect on any cytochrome P450 enzymes. Promisingly, chicoric acid, quinine, curcumin, and demothoxycurcumin exhibited high binding affinity on SARS-CoV-2 target proteins and expressed good drug-likeness and pharmacokinetic properties. Further research is required to investigate the potential uses of these compounds in the treatment of SARS-CoV-2.


Introduction
Coronavirus disease 2019 (COVID- 19), the highly infectious viral illness caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), has had a disastrous effect as relieving cytokines storm syndrome, protecting human tissues and organs, relieving immunological injury, and enhancing the body's ability [17].
In this study we aim to investigate docking characteristics of 10 bioactive compounds identified from traditional medicines (caffeic and chicoric acids from Echinacea purpurea; xanthorrhizol from Java turmeric; curcumin and demothoxycurcumin from Curcuma longa L.; Quinine from Chinchona bark; Theaflavin from Black tea; withaferin A and withanolide D from Withania somnifera L.; cepharanthine from Stephania cepharantha) on S-protein, RdRp and SARS-CoV-2 PLpro to explore the potential mode of the inhibitory activity toward SARS-CoV-2. The molecules that showed better hydrogen bonding and binding energy were selected further for drug likeness and pharmacokinetics prediction.

Results and Discussion
Herbal medicines and medicinal plant-based natural compounds provide a rich resource for novel antiviral drug development. Some natural medicines have been shown to possess antiviral activities against various virus strains including coronavirus, herpes simplex virus [23][24][25], influenza virus [25], human immunodeficiency virus [26], hepatitis B and C viruses [27], SARS, and MERS [28,29]. These compounds antiviral action mechanisms caused by influencing of the viral life cycle, such as viral entry, replication, assembly, and release, as well as virus-host-specific interactions. In this study we analyze the binding mode of 10 ligands Figure 1 that previously identified in some traditional plants Table 1 using computational docking analysis. All tested ligands showed hydrogen interaction to papain-like protease (6W9C) except cepharanthine and theaflavin. The ligand docking on CYS111, HIS272, and ASP286 catalytic residues ( Figure 2A) ranked by binding energy ∆G in kcal/mol were as follows: chicoric acid > demothoxycurcumin > curcumin > withanolide D > xanthorrhizol > withaferin A > caffeic acid > quinine. It is known that the high binding affinity of the drug compounds depends on the type and amount of bonds occurring with the target protein [53]. The binding energies were −7.62, −6.81, −6.70, −6.58, −6.16, −6.13, −5.70, and −5.67 kcal/mol, respectively (Table 2; Figure 3).
In 2021 a Japanese research team reported the anti-COVID-19 activity of the combining cepharanthine/nelfinavir providing synergistic antiviral effects that CEP and nelfinavir inhibit SARS-CoV-2 entry and RNA replication, respectively [58]. In 2021, a Japanese research team reported that the combination of cepharanthine and nelfinavir inhibited SARS-CoV-2 entrance and RNA replication, offering synergistic antiviral effects [58]. flammatory, antiparasitic, anti-oxidative, antiviral, and anti-HIV activities [32][33][34][35]. In 2021 a Japanese research team reported the anti-COVID-19 activity of the combining cepharanthine/nelfinavir providing synergistic antiviral effects that CEP and nelfinavir inhibit SARS-CoV-2 entry and RNA replication, respectively [58]. In 2021, a Japanese research team reported that the combination of cepharanthine and nelfinavir inhibited SARS-CoV-2 entrance and RNA replication, offering synergistic antiviral effects [58].  This previous combination based on in silico docking simulation that confirms that CEP molecule can bind to SARS-CoV-2 S-protein and interfere with the spike engagement to its receptor, angiotensin-converting enzyme 2 (ACE2) [59,60]. Our results strongly supported this mechanism of CEP toward SARS-CoV-2 S-protein. However, chicoric acid, quinine, withaferin A, withanolide D, theaflavin, demothoxycurcumin, curcumin, and xanthorrhizol expressed higher binding energy than cepharanthine in case SARS-CoV-2 S-protein binding. From all of this we supposed that chicoric acid and quinine may play the same role in this combination (cepharanthine/nelfinavir) beside their acting mechanism toward SARS-CoV-2, PLpro and RdRp that may be promising to limit SARS-CoV-2 proliferation than CEP.
Regarding to RdRp (6M71), all tested ligands showed H-interaction except Cepharanthine. The ligand docking on ARG 553, ARG 555, LYS 545, and ASN 691 ( Figure  2C) ranked by binding energy ΔG in kcal/mol were as follows: curcumin = quinine > demothoxycurcumin > chicoric acid > caffeic acid > withaferin A > withanolide D > xanthorrhizol > theaflavin. The binding energies were −7.80, −7.80, −7.64, −7.50, −6.75, −6.27, −6.10, −5.33, and −2.55 kcal/mol, respectively (Table 2; Figure 5). Curcumin (CC) and its analogues are the main phytonutrients of turmeric (Curcuma longa L.) and other Curcuma spp., which are widely used around the world as culinary Curcumin (CC) and its analogues are the main phytonutrients of turmeric (Curcuma longa L.) and other Curcuma spp., which are widely used around the world as culinary spices, traditional medicine as well as a popular dietary supplement ingredient due to its wide range of health benefits including anti-inflammation [61], anti-cancer [62], cardiovascular regulation [63], respiratory [64], and immune system benefits [65]. In addition, the suppression of multiple cytokines by curcumin suggested that it may be a useful approach in treating Ebola patients against cytokine storm [66]. As severe cases of COVID-19 are often associated with cytokine release syndrome, the use of anti-inflammatory molecules may reduce the proinflammatory cytokines involved. Quinine, is an extract of the bark of the Chinchona tree (native to the Andes of South America), that was used to treat feverish infections, particularly malaria, for hundreds of years almost worldwide [67]. Curcuma longa L. and Chinchona bark are rich source of curcumin and quinine respectively (Table 1). Table 3 presents the drug likeness and pharmacokinetic properties of the tested ligands. Lipinski's Rule of Five is generally used as an indicator of the drug likeness and pharmacological activities. In humans, this would make them orally active medications [68]. All the tested compounds are in the molecular weight range of 180.16 to 474.37 Da (<500 Da), except for cepharanthine and theaflavin are 606.71 and 564.49 Da. It is well known that drug molecules typically have low molecular weight (<500 kDa) are transported, diffused, and absorbed easily compared with large molecules [69]. All the tested compounds also have fewer than 15 rotatable bonds, and all have less than 5 hydrogen bond donors (NH and OH), except for quinine and theaflavin. In addition, the numbers of hydrogen bond acceptors (O and N atoms) predicted in all compounds are less than 10, except for chicoric acid and theaflavin (Table 3). At the same time, permeability (logP) of these ligands has also been investigated, and it was found that these ligands exhibited logP values of less than 5 except cepharanthine. Moreover, the topological polar surface area (TPSA) values of all ligands are less than 140 Å, except for chicoric acid and theaflavin. TPSA is closely related to the H-bonding ability of a compound [70]. The TPSA and the logP values are the two essential characteristics in the analysis of the bioavailability of drug molecules and permeability toward bio-membranes [71]. Compounds with 10 rotatable bonds and TPSA of ≤140 Å can be predicted to have good bioavailability [72]. As a result, all of the ligands examined have good bioavailability, with the exception of chicoric acid and theaflavin, which have TPSA values of 208.12 and 217.60, respectively. Pharmacokinetic prediction provides information to evaluate the time course of drugs and their effects on the body to design an appropriate drug regimen for any patient [73]. Our results of pharmacokinetic prediction (Table 3) showed that all tested ligands demonstrate a high absorption rate in the GI tract, except for chicoric acid and theaflavin. Moreover, none of the tested ligands were predicted to be able to pass through the bloodbrain barrier, except for quinine and xanthorrhizol. Additionally, chicoric acid, withaferin A and withanolide D are the only compounds predicted to act as a substrate for P-gp, which decrease their further clinical application [74]. The predicted pharmacokinetic behavior indicates that only caffeic acid, cepharanthine, chicoric acid, withaferin A, and withanolide D would have no inhibitory effect on any cytochrome P450 enzymes (CYP1A2, CYP2C19, CYP2D6, and CYP3A4). Furthermore, CYP2C9 activity is only known to be affected by curcumin, demothoxycurcumin, theaflavin and xanthorrhizol. Cytochrome P450 enzymes are essential for the metabolism of many chemicals. Although this class contains more than 50 enzymes, six of them metabolize 90% of compounds. Cytochrome P450 enzyme inhibition causes unwanted adverse effects or therapeutic failures [75]. The proposed mechanism of SARS-CoV-2 inhibition by tested compounds is summarized in Figure 6; chicoric acid, quinine and withaferin A, interacting with S protein, inhibits viral ability to attach to human ACE2. Additionally, chicoric acid, demothoxycurcumin, and curcumin interacting with PL pro, inhibits viral replication, moreover, curcumin, demothoxycurcumin, and quinine interacting with RdRp, inhibits viral replication. Furthermore, interferon (IFN) production is triggered by virus infections that cause accumulation of RNA species in the cytoplasm, initiate the oligomerization of an essential mitochondrial antiviral signaling protein, MAVS, that serves as a scaffold for the activation of TNF receptor-associated factor (TRAF) family proteins, TRAF2, TRAF3, TRAF5, and TRAF6. TRAF proteins catalyze the assembly of K63-linked ubiquitin chains that are required for the activation of serine kinases, IKKa, IKKb, IKKc, IKKe, and TBK1. Kinase activation triggers the phosphorylation and nuclear import of IRF3, driving the production of primary antiviral effectors including IFNs [76]. On the other hand, PLpro of SARS-CoV-2 interfere with innate immune response by directly cleaving IRF3, inhibits IFNβ production. In general, in silico studies are a great starting point when looking for new therapeutic targets. In the event of an unpredictable pandemic, such as the recent COVID-19 epidemic, in silico research are critical. Drug targets cannot be discovered in a short period of time using existing methods. To reach bulk targets, it is unavoidable to have extremely rapid procedures. In silico approaches can be used to speed up the search for drug targets in natural medicinal plants. This article's results warrant further evaluation of the potential anti-SARS-CoV-2 activity of these ligands in vitro.
Molecules 2021, 26, x 12 of 16 activation triggers the phosphorylation and nuclear import of IRF3, driving the production of primary antiviral effectors including IFNs [74]. In blue: PLpro of SARS-CoV-2 interfere with innate immune response by directly cleaving IRF3, inhibits IFNβ production. Figure 6. The proposed mechanism of SARS-CoV-2 inhibition by tested compounds.

Docked SARS-CoV-2 Protein Structures
The structure of the SARS-CoV-2 papain-like protease (6W9C), RNA-dependent RNA polymerase (6M71), and spike protein (6M17) used as a target for ligands binding was downloaded from RCSB website [77]. PDB (Protein Data Bank) has enabled breakthroughs in research, such as this study, and education worldwide [78].

Ligands and Drug Scan
The three-dimensional (3D) structures of all tested ligands were drawn in ACD/ChemSketch and then docked into the rigid binding pocket of 6W9C, 6M71, and 6M17 of SARS CoV-2. The compounds used in the present study were caffeic acid, Steps from one to six describe the replication cycle of SARS-CoV-2; attachment (1), viral entry (2), translation of viral proteins (3), viral replication (4), assembly (5), and release (6). In red: chicoric acid, quinine, and withaferin A, interacting with S protein, inhibiting its viral ability to attach to human ACE2 (1). Additionally, chicoric acid, demothoxycurcumin, and curcumin interacting with PL pro inhibits viral replication (2), moreover, curcumin, demothoxycurcumin, and quinine interacting with RdRp inhibits viral replication (3). Interferon (IFN) production is triggered by virus infections that cause accumulation of RNA species in the cytoplasm, initiating the oligomerization of an essential mitochondrial antiviral signaling protein, MAVS, which serves as a scaffold for the activation of TNF receptor-associated factor (TRAF) family proteins, TRAF2, TRAF3, TRAF5, and TRAF6. TRAF proteins catalyze the assembly of K63-linked ubiquitin chains that are required for the activation of serine kinases, IKKa, IKKb, IKKc, IKKe, and TBK1. Kinase activation triggers the phosphorylation and nuclear import of IRF3, driving the production of primary antiviral effectors including IFNs [74]. In blue: PLpro of SARS-CoV-2 interfere with innate immune response by directly cleaving IRF3, inhibits IFNβ production.

Docked SARS-CoV-2 Protein Structures
The structure of the SARS-CoV-2 papain-like protease (6W9C), RNA-dependent RNA polymerase (6M71), and spike protein (6M17) used as a target for ligands binding was downloaded from RCSB website [77]. PDB (Protein Data Bank) has enabled breakthroughs in research, such as this study, and education worldwide [78].

Ligands and Drug Scan
The three-dimensional (3D) structures of all tested ligands were drawn in ACD/Chem-Sketch and then docked into the rigid binding pocket of 6W9C, 6M71, and 6M17 of SARS CoV-2. The compounds used in the present study were caffeic acid, cepharanthine, chicoric acid, curcumin, demothoxycurcumin, quinine, theaflavin, withaferin a, withanolide d and xanthorrhizol. The drug likeness and pharmacokinetic properties were calculated using the SWISSADME prediction website (http://www.swissadme.ch/) (accessed on 15 December 2021) [68,79]. Figure 1 shows the structure of the 10 docked ligands we used herein. Table 4 contains the amino acids binding hits of the SARS-CoV-2 polymerase as described by Afonine et al. [80], papain-like protease (6W9C) as described by Baez-Santos et al., 2015 and spike protein (6M17) as described by Baig et al. [7].

Molecular Docking
Ligand optimization was performed using Open Babel, converting ligands from mol into the PDB format. Autodock version 4.0 was used for protein optimization through the removal of water and other atoms and then addition of a polar hydrogen group. Ligand tethering of the protein was performed by regulating the genetic algorithm (GA) parameters using 10 runs of the GA criteria. Docking analyses and determination of hydrogen bonds (H-bonds) were conducted using Chimera 1.8.1 [81].

Conclusions
Antiviral treatments targeting the coronavirus disease 2019 are urgently required. We screened a panel of ligands from traditional plant sources already used in medicinal fields and some of them described for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2). Chicoric acid, quinine, curcumin, and demothoxycurcumin exhibited high binding affinity on SARS-CoV-2 S-protein, PLpro, and RdRp. Binding of these proteins interfere with the viral entry, replication, and immune response evasion. Therefore, these compounds may have a great potential for inhibiting the virus. Our work provides data about the ligand mechanism toward SARS-CoV-2 target proteins and allows the comparison between them based on scientific information. Further in vitro cell-based investigations will be needed for chicoric acid, quinine, curcumin, and demothoxycurcumin compounds to confirm their previous results and determine their antiviral mechanism at cell level.