In Silico Screening of Potential Phytocompounds from Several Herbs against SARS-CoV-2 Indian Delta Variant B.1.617.2 to Inhibit the Spike Glycoprotein Trimer

: In October 2020, the SARS-CoV-2 B.1.617 lineage was discovered in India. It has since become a prominent variant in several Indian regions and 156 countries, including the United States of America. The lineage B.1.617.2 is termed the delta variant, harboring diverse spike mutations in the N-terminal domain (NTD) and the receptor-binding domain (RBD), which may heighten its immune evasion potentiality and cause it to be more transmissible than other variants. As a result, it has sparked substantial scientiﬁc investigation into the development of effective vaccinations and anti-viral drugs. Several efforts have been made to examine ancient medicinal herbs known for their health beneﬁts and immune-boosting action against SARS-CoV-2, including repurposing existing FDA-approved anti-viral drugs. No efﬁcient anti-viral drugs are available against the SARS-CoV-2 Indian delta variant B.1.617.2. In this study, efforts were made to shed light on the potential of 603 phytocompounds from 22 plant species to inhibit the Indian delta variant B.1.617.2. We also compared these compounds with the standard drug ceftriaxone, which was already suggested as a beneﬁcial drug in COVID-19 treatment; these compounds were compared with other FDA-approved drugs: remdesivir, chloroquine, hydroxy-chloroquine, lopinavir, and ritonavir. From the analysis, the identiﬁed phytocompounds acteoside ( − 7.3 kcal/mol) and verbascoside ( − 7.1 kcal/mol), from the plants Clerodendrum serratum and Houttuynia cordata , evidenced a strong inhibitory effect against the mutated NTD (MT-NTD). In addition, the phytocompounds kanzonol V ( − 6.8 kcal/mol), progeldanamycin ( − 6.4 kcal/mol), and rhodoxanthin ( − 7.5 kcal/mol), from the plant Houttuynia cordata , manifested signiﬁcant prohibition against RBD. Nevertheless, the standard drug, ceftriaxone, signals less inhibitory effect against MT-NTD and RBD with binding afﬁnities of − 6.3 kcal/mol and − 6.5 kcal/mol, respectively. In this study, we also emphasized the pharmacological properties of the plants, which contain the screened phytocompounds. Our research could be used as a lead for future drug design to develop anti-viral drugs, as well as for preening the Siddha formulation to control the Indian delta variant B.1.617.2 and other future SARS-CoV-2 variants.


Introduction
Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) first emerged in December 2019 and has since spread to over 221 nations, leading the continuing outbreak to be declared a worldwide medical emergency [1,2]. As of 17 December 2021, there have been more than 271.96 million confirmed cases and about 5.33 million deaths reported across almost 200 countries [3]. The treatment for individuals infected with SARS-CoV-2 naturally occurring chemicals employing in silico techniques might be promising treatments for COVID-19, achieved within a short timeframe and at a reasonable price [48][49][50][51][52][53][54][55][56].
Siddha and Ayurvedic medicine, which originated in Tamil Nadu, South India, are two of the oldest traditional remedies [57][58][59]. The government of India has recommended Kabasura kudineer choornam for combating a COVID-19 viral infection. Since the lungs are the primary organ of kapha, Kabasura kudineer choornam is a polyherbal Siddha composition that includes 15 herbs suggested to manage typical respiratory illnesses, such as colds, coughs, breathing difficulties, and the flu [60][61][62]. In silico investigations revealed that the Kabasura kudineer formulation had anti-inflammatory, antipyretic, and antibacterial properties; as well, it poses a better ability to bind to the SARS-CoV-2 spike protein [63,64]. Likewise, Houttuynia cordata (H. cordata) played a vital role in herbal therapy for the SARS outbreak in Southern China in 2003 [65]. Furthermore, it acts effectively against chikungunya, human noroviruses, human herpes viruses, the dengue virus, influenza, the pseudorabies virus, and murine coronaviruses [66,67]. Das et al. discovered H. Cordata phytocompounds to be a potential inhibitor for Mpro and PLpro, thereby preventing the replication of SARS-CoV-2 [68]. Similarly, the dried flower bud Syzygium aromaticum (S. aromaticum), a plant indicated by the English name "clove," acts effectively against COVID-19 [69]. Spices such as cloves are used in three forms: whole dried buds, powdered cloves, or extracted as an essential oil. According to earlier literature research, additional noteworthy qualities include the ability to treat colds, cough, asthma, and upper respiratory diseases, as well as anti-cancer, anti-inflammatory, and antimutagenic activity [70]. Tallei et al. reported that the phytocompounds hesperidin, nabiximols, pectolinarin, epigallocatechin gallate, and rhoifolin from Citrus spp. are effective against the Mpro and the spike glycoprotein trimmer (S-Protein) of SARS-CoV-2, which inhibits proliferation of the virus [71]. According to Khazdair et al., Nigella sativa (N. sativa) has protective effects on obstructive lung disorders, and this herb might be helpful in the treatment of COVID-19 [72].
The phytocompounds of N. sativa impede viral entrance and reproduction within the host cell by interfering with its binding to ACE2 receptors [73,74]. Interestingly, Shree et al. stated that three phytoconstituents from Ocimum basilicum (O. basilicum) vicenin, sorientin 4 -O-glucoside 2"-O-p-hydroxy-benzoagte, and ursolic acid inhibited the Mpro of SARS-CoV2 [75]. Likewise, consuming Piper nigrum (black pepper) (P. nigrum) or piperine may help limit viral growth [76,77]. The Ministry of AYUSH, Government of India, also described that black pepper might have an anti-SARS-CoV-2 role [78]. Metastasio et al. concluded from their study that short-term kratom usage might reduce pain associated with COVID-19 infection without causing physical or psychological withdrawal symptoms when the kratom was halted [79]. Therefore, considering the therapeutic importance of Kabasura kudineer, H. cordata, S. aromaticum, Citrus spp., N. sativa, O. basilicum, P. nigrum, and Mitragyna speciosa Korthi (M. speciosa Korthi) and their strong ethnopharmacological background, the present study is primarily intended to perform molecular docking studies with these crucial phytocompounds acting against the S-protein of the SARS-CoV-2 Indian delta variant B.1.617.2.

Target Preparation and Ligand Library
The cryo-electron microscopy structure of the S-Protein, a subunit vaccine candidate for COVID-19 PDB: 7E7B [87], was downloaded from the protein data bank and edited to remove unnecessarily bounded ligands and water molecules using the Discovery Studio Visualizer v19.1.0.18287 (www.accelerys.com) (accessed on 4 November 2021) and saved in PDB format. The major phytoconstituents present in Kabasura kudineer choornam and the other selected herbs were retrieved in SDF file format, and some compounds in 2D structures were also obtained from the PubChem database. A 3D structure was delineated for each of the obtained 2D structures and optimized with a force field based on Chemistry at Harvard Macromolecular Mechanics (CHARMM) parameterization using ACD/Chemsketch vC05E41 (Advanced Chemistry Development, Inc., Toronto, ON, Canada). The 3D structures were saved in the SDF file format, and all the obtained phytocompounds were then converted into PDB file format using OPEN BABEL software [88].

Mutated NTD Model
In this study, the NTD domain from PDB: 7E7B was mutated with Asn165Gln and Asn234Gln. No significant conformational change was observed in the structure of the mutant NTD model. Structural evaluation with RAMPAGE showed similar residue numbers in the most favored region (96.81%). Structural alignment and superimposition of wild type (WT-NTD) and mutant type (MT-NTD) (Asn165Gln and Asn234Gln) were performed with the 3D-SS server (http://cluster.physics.iisc.ernet.in/3dss/) (accessed on 26 October 2021) to calculate the disparity in the mutated sites. The superimposed structure was visualized in the Discovery Studio Visualizer v19.1.0.18287 software.

Molecular Docking
After preparing the phytocompounds as ligands and the receptor RBD and MT-NTD from 7E7B as targets, PyRx was implied with the Autodock Vina option using the new scoring function [89]. It analyzes the docking propensity and interfaces between the ligands, RBD, and mutated NTD. The prepared targets and ligands were converted into a PDBQT file format. For our docking analysis, we applied the specific search anchoring function of the PyRxVirtual Screening tool. The grid box properties were set as size_x = 31.43 Å, y = 46.11 Å, and size_z = 32.15 Å for the NTD molecular docking and size_x = 36.38 Å, size_y = 67.04 Å, and size_z = 32.08 Å was set for the RBD molecular docking and then docked. The ligands were screened out for a binding affinity of ≤6.0 kcal/mol. The significant interaction between the ligands and the receptors' binding site was acquired in 2D and 3D formats by importing the docked results into the LigPlot+, PyMol, and Discovery Studio Visualizer v19.1.0.18287 (www.accelerys.com) (accessed on 4 November 2021). In the autodock vina scoring function, where C is sum of intermolecular and intramolecular distance; ∑ is the overall pairs of atoms; f titj is symmetric set of interaction functions; and r ij is interatomic distance.

Evaluation of Ligands Drug Likeness and Toxicity
The screened ligands were evaluated for draggability, physicochemical properties, toxicity, toxicity classes, and lethal dose using the Molinspiration server (www.molinspiration. com/cgi-bin/properties) (accessed on 4 November 2021). The druggability properties were analyzed based on the molar weights (MW), total polar surface area (TPSA), lipophilicity (log P), hydrogen bond acceptor (HBA), and hydrogen bond donor (HBD) to identify Lipinski's rule of the drug-like compounds. In addition, the simplified molecular-input line-entry system (SMILES) was downloaded from the PubChem Database to calculate the ADMET properties with toxicity class. The ADMET properties were calculated by implementing ADMETlab 2.0 [83] and ProTox-II with default parameters [84].

Results
The cryo-electron microscopy structure of the S-Protein trimer, a subunit vaccine candidate for COVID-19 PDB: 7E7B [87], is depicted in Figure 1a, containing chain A, chain B and chain C. Chain A was separated from PDB, and the 7E7B S-Protein trimmer contains the three major units: NTD (14-305), RBD (329-521), and S2 subunit (522-1147), as described in Figure 1b. The structure of the MT-NTD and the mutation in the RBD are pictured in Figure 1c.
The MT-NTD structure was compared with the WT-NTD, and the structural variation and coordination were evaluated. The WT-NTD ( Figure 2a) and MT-NTD ( Figure 2b) were validated using the Ramachandran plot. For the WT-NTD and MT-NTD, 98.92% residues were present in the most favored region, and no significant changes were noted between the WT-NTD and MT-NTD. The RMSD and sequence identity were compared for the superimposed structure of the MT-NTD with WT-NTD. The RMSD and sequence identity of the MT-NTD was 0.004 Å, and 99.68%, while the WT-NTD was set as a fixed molecule. Furthermore, the stamp score was noted as 9.799 for the MT-NTD out of 10. The superimposed WT-NTD and MT-NTD structures' stamp sequence alignment is shown in Figure S1 (Supplementary File). The change in two amino acids did not produce any considerable alteration in the overall structural conformation of the protein in terms of the smaller RMSD and sequence identity. The superimposition of binding residues for the WT-NTD (Asn165, Asn234) and MT-NTD (Gln165, Gln234) models are shown in Figure 2c, respectively.
We used 603 phytocompounds in the screening process, obtained from the 22 wellannotated herbal plants included with the Kabasura kudineer choornam (Figure 3a). After processed molecular docking, the ligands with a higher binding affinity (≤6.0 kcal/mol) were screened from the docked results ( Figure 3b   We used 603 phytocompounds in the screening process, obtained from the 22 wellannotated herbal plants included with the Kabasura kudineer choornam (Figure 3a). After processed molecular docking, the ligands with a higher binding affinity (≤6.0 kcal/mol) were screened from the docked results ( Figure 3b).  The chemical properties, including the molecular formula, molecular weight, and PubChem ID for the phytocompounds acted significantly against the NTD and RBD (binding affinity ≤ −6.0 kcal/mol), listed in Tables S1-S4 (Supplementary File S1). Th phytocompounds that beneficially interacted with the mutated NTD domain, its binding affinity, and LigPlot interactions are listed in Table 1 and Figure S2 (Supplementary Fil S1). The chemical properties, including the molecular formula, molecular weight, and PubChem ID for the phytocompounds acted significantly against the NTD and RBD (binding affinity ≤ −6.0 kcal/mol), listed in Tables S1-S4 (Supplementary File S1). The phytocompounds that beneficially interacted with the mutated NTD domain, its binding affinity, and LigPlot interactions are listed in Table 1 and Figure S2 (Supplementary File S1). The screened phytocompounds binding affinity and its LigPlot interactions for the RBD domain are listed in Table 2 and Figure S3 (Supplementary File S1). The pharmacological activity of the resulting active compounds acts against the NTD and RBD is listed in Table 3, and the predicted drug-likeness and toxicity classes are presented in Tables 4 and 5.    We finally identified the compounds acteoside and verbascoside from C. serratum as acting effectively against the mutated NTD ( Figure 4D-I,K,L). The acteoside demonstrated more beneficial interactions with PHE86, ASN87, ASP88, ASN196, ASP198, GLY199, ILE233, and GLN234 via hydrogen bonds with a binding affinity of −7.3 kcal/mol. It also holds alkyl interactions with the ILE197. It is surrounded by hydrophobic residues (LEU54, LYS195, THR236). The toxicity class was also identified as V. The verbascoside interacts with ASN87, ASP88*, ASN196, ASP198, GLY199, ILE233, and GLN234 via carbon-hydrogen* bond with a binding affinity of −7.1 kcal/mol. It also held pi-alkyl interaction with LEU54 and PRO272, as well as pi-sigma bonding with THR236, and was surrounded by the hydrophobic residues (GLN52, PRO85, and ILE235). The predicted toxicity class of verbascoside was V. The standard drug ceftriaxone interacted with ASP88, ASN87, GLN115, ASN156*, ASP198, GLY199, GLY232, GLN234*, and ILE233 by hydrogen, carbonhydrogen, and pi-alkyl bonds. It was also surrounded by hydrophobic residues ASN196 and ILE197, and the predicted toxicity class was VI. Even though the standard drug ceftriaxone interacted with the crucial residues GLN234* by carbon-hydrogen bond, the binding affinity was −6.3 kcal/mol. Moreover, the compounds kanzonol V, progeldanamycin, and rhodoxanthin from H. cordata excellently inhibit the RBD domain compared with the other tested phytocompounds ( Figure 5D-L,N-P). Kanzonol V strongly interacted with GLU484 and TYR449 by hydrogen bond and showed pi-alkyl interactions with PHE456 and TYR489. Kanzonol V is surrounded by hydrophobic residues LEU452, LEU455, PHE490, and LEU492. The binding affinity is −6.8 kcal/mol. It belonged to the toxicity classes of V. Progeldanamycin has hydrogen bond interactions with TYR449, and SER494, as well as pi-sigma interaction with LEU452, and is also surrounded by the hydrophobic residues ASN450, PHE490, LEU492, and GLN493. The observed binding affinity is −6.4 kcal/mol with the toxicity class IV. Rhodoxanthin shows pi-alkyl interactions with LEU452, TYR489, and PHE490 and was surrounded by the hydrophobic residues ARG346, SER349, TYR351, ASN450, LEU455, PHE456, GLU484, GLY485, PHE486, LEU492, GLN493, and SER494 with the binding affinity of −7.5 kcal/mol. The toxicity class was noticed as VI. However, the ceftriaxone did not interact with the specifically defined residues L452R, T478K, and E484Q in the Indian delta variant, and the binding affinity was observed as −6.5 kcal/mol.

Discussion
The altered viral subset will become predominant in the context of environmental switches. In the face of such problems in treatment planning, there is now strong necessity for alternative therapy. Natural substances, which have been used globally for many years due to their chemical variety, species diversity, and drug-like features, may be recognized in this category. In this line, we aimed to assess the anti-COVID-19 capability of Siddha polyherbal formulation and plant extracts containing different phytocompounds that have been previously reported to have antiviral properties [54,55]. The SARS-spike CoV-2 protein has undergone mutations and is heavily glycosylated; hence, the biological relevance of viral alterations must be investigated promptly [40,131]. Li et al. examined over 80 variations and 26 glycosylation site variations to identify the severity of disease transmission and sensitivity from recovered patients. The mutated residue N234Q was significantly resistant to neutralizing antibodies, whereas N165Q became more susceptible [48]. Similarly, the Indian delta variant B.1.617.2. carries five particular mutations: L452R, T478K, E484Q, D614G, and P681R [12,20,132]. It controls viral fitness by increasing the ACE2 receptor binding affinity, increasing infectivity, and deactivating the antibodies. These mutations in the Indian delta variant increased the spike's stability, viral infectivity, and stronger cell attachment, thereby promoting viral replication, transmissibility, and pathogenicity [12,29,133,134]. These discoveries could help with vaccine and therapeutic antibody development [40].
Drugs derived from medicinal plants have traditionally been widely utilized to treat diseases [135][136][137][138]. However, effective medication is strongly advocated at this key stage of COVID-19 and its variant infection [1]. Plants are rich in phytocompounds, which could efficiently counteract with COVID-19 [139,140]. Hence, the MT-NTD and RBD domains in the SARS-CoV-2 were docked using 603 phytochemicals in this investigation. Many of the phytocompounds found in plants had high protein-binding abilities. After applying the threshold criteria of ≤ −6.0 kcal/mol to 603 compounds, 27 and 13 phytocompounds for the MT-NTD and RBD, respectively, were screened.
Phytocompounds with a binding affinity of ≤ −6.0 kcal/mol were chosen because they showed promising inhibition with crucial residues N234Q, N165Q, L452R, T478K, and E484Q against the MT-NTD and RBD. Typically, the threshold is chosen by correlating it to known inhibitors that have been previously published and biologically verified. In this regard, we selected ceftriaxone, a standard drug that evidenced higher binding affinity

Discussion
The altered viral subset will become predominant in the context of environmental switches. In the face of such problems in treatment planning, there is now strong necessity for alternative therapy. Natural substances, which have been used globally for many years due to their chemical variety, species diversity, and drug-like features, may be recognized in this category. In this line, we aimed to assess the anti-COVID-19 capability of Siddha polyherbal formulation and plant extracts containing different phytocompounds that have been previously reported to have antiviral properties [54,55]. The SARS-spike CoV-2 protein has undergone mutations and is heavily glycosylated; hence, the biological relevance of viral alterations must be investigated promptly [40,131]. Li et al. examined over 80 variations and 26 glycosylation site variations to identify the severity of disease transmission and sensitivity from recovered patients. The mutated residue N234Q was significantly resistant to neutralizing antibodies, whereas N165Q became more susceptible [48]. Similarly, the Indian delta variant B.1.617.2. carries five particular mutations: L452R, T478K, E484Q, D614G, and P681R [12,20,132]. It controls viral fitness by increasing the ACE2 receptor binding affinity, increasing infectivity, and deactivating the antibodies. These mutations in the Indian delta variant increased the spike's stability, viral infectivity, and stronger cell attachment, thereby promoting viral replication, transmissibility, and pathogenicity [12,29,133,134]. These discoveries could help with vaccine and therapeutic antibody development [40].
Drugs derived from medicinal plants have traditionally been widely utilized to treat diseases [135][136][137][138]. However, effective medication is strongly advocated at this key stage of COVID-19 and its variant infection [1]. Plants are rich in phytocompounds, which could efficiently counteract with COVID-19 [139,140]. Hence, the MT-NTD and RBD domains in the SARS-CoV-2 were docked using 603 phytochemicals in this investigation. Many of the phytocompounds found in plants had high protein-binding abilities. After applying the threshold criteria of ≤−6.0 kcal/mol to 603 compounds, 27 and 13 phytocompounds for the MT-NTD and RBD, respectively, were screened.
Phytocompounds with a binding affinity of ≤−6.0 kcal/mol were chosen because they showed promising inhibition with crucial residues N234Q, N165Q, L452R, T478K, and E484Q against the MT-NTD and RBD. Typically, the threshold is chosen by correlating it to known inhibitors that have been previously published and biologically verified. In this regard, we selected ceftriaxone, a standard drug that evidenced higher binding affinity toward both the NTD and RBD than the FDA-approved anti-viral drugs lopinavir [141], remdesivir [142], chloroquine [143], umifenovir [144], favipiravir [145], ribavirin [146], hydroxychloroquine [147], sofosbuvir [148], and oseltamivir [4]. However, ceftriaxone shows less binding affinity at −6.3 kcal/mol and −6.5 kcal/mol towards the MT-NTD and RBD domains. Likewise, there is no interaction with the specified residues accounting for the targeted MT-NTD and RBD of the S-Protein of the Indian delta variant B.1.617.2.
Nevertheless, the ceftriaxone constitutes less binding affinity −6.5 kcal/mol, and it does not interact with the targeted residues L452R, T478K, and E484Q in the RBD. The compounds kanzonol V (−6.8 kcal/mol), progeldanamycin (−6.4 kcal/mol), and rhodoxanthin (−7.5 kcal/mol) from H. cordata act effectively against the target residues in the RBD of the Indian delta variant B.1.617.2 with better binding affinity. Even though kanzonol and rhodoxanthin violated one or two Lipinski rules, they beneficially treated bacterial and fungal infections, as previously reported [151]. Additionally, the toxicity class is satisfactorily categorized as V and VI for kanzonol V and rhodoxanthin. Progeldanamycin is not violating any rules, and its toxicity class is described as VI. In addition, Benet et al. explained that violation of two or more Lipinski rules for the natural products and their derivatives is acceptable. However, the FDA-approved oral drugs under the category Class 1 (acarbose, cyanocobalamin, everolimus, ivermectin, etc.) are believed to have high solubility properties and high permeability, also violating the Lipinski rule [152].
Moreover, the maximum number of phytocompounds from H. cordata and Kabasura kudineer act significantly well against the NTD and RBD domain of the Indian delta variant B.1.617.2 S-protein. AYUSH has recommended many treatments towards COVID-19 prevention, which are implemented as preventative and symptomatic therapy in COVID-19 management [153,154]. Nevertheless, no Siddha formulation was prescribed for COVID-19 containing H. cordata, including Kabasura kudineer and Ayush kwath, which has already been recommended by the Indian government [60,155]. According to the current investigation, the formulations and phytocompounds examined in this study showed considerably greater binding efficacy against the MT-NTD and RBD in the S-Protein of the Indian delta variant B.1.617.2. In silico studies suggested that the resulting phytocompounds may operate as efficient inhibitors of the Indian delta variant B.1.617.2 by binding to the spike glycoprotein, which may be investigated further in vitro to develop improved herbal formulations and anti-viral drugs. Furthermore, SARS-CoV-2 has been proven to have a greater affinity for pharyngeal epithelial cells [55]. Since these extracts can be delivered to the pharyngeal regions through appropriate oral formulations, they will be effective to control the infection rates.

Conclusions
During the COVID-19 disease outbreak caused by SARS-CoV-19 and its variants the disease transmission heightened due to a lack of targeted medications and vaccines. Even though vaccines have been identified, their efficacy against the Indian delta variant B.1.617.2 has dramatically decreased, forcing researchers to look for novel anti-viral formulations. In this regard, we analyzed 603 compounds from 22 plants. We identified five compounds: acteoside, verbascoside, kanzonol V, progeldanamycin, and rhodoxanthin, which acted significantly against the Indian delta variant B.1.617.2 compared with ceftriaxone, which is the most beneficial drug in COVID 19 treatment. Though the Siddha formulation contains C. serratum (L.) Moon, there is no Siddha formulation containing H. cordata. Hence, this study contributes to the evidence for developing pharmaceutical formulations and antiviral drugs that act specifically against the Indian delta variant B.1.617.2.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/ 10.3390/app12020665/s1, Figure S1: The sequence alignment for the wild-and mutated-type NTD in S-Protein; Figure S2: The LigPlot interaction for the mutated NTD domain and the screened phytocompounds based on the binding affinity ≤−6.0 kcal/mol; Figure S3: The LigPlot interaction for the RBD domain and the screened phytocompounds based on the binding affinity ≤−6.0 kcal/mol. Table S1: The molecular properties of screened phytocompounds with the ≤−6.0 kcal/mol binding energy against mutated NTD Domain; Table S2: The molecular properties of screened phytocompounds with the ≤−6.0 kcal/mol binding energy against RBD Domain; Table S3: The binding affinity, interacting residues, and bond length for the screened phytocompounds against the mutated NTD in the S-Protein; Table S4: The binding affinity, interacting residues, and bond length for the screened phytocompounds against the RBD in the S-Protein.

Data Availability Statement:
The data presented in this study are available within the article.