An Investigation of the Antiviral Potential of Phytocompounds against Avian Infectious Bronchitis Virus through Template-Based Molecular Docking and Molecular Dynamics Simulation Analysis

Vaccination is widely used to control Infectious Bronchitis in poultry; however, the limited cross-protection and safety issues associated with these vaccines can lead to vaccination failures. Keeping these limitations in mind, the current study explored the antiviral potential of phytocompounds against the Infectious Bronchitis virus using in silico approaches. A total of 1300 phytocompounds derived from fourteen botanicals were screened for their potential ability to inhibit the main protease, papain-like protease or RNA-dependent RNA–polymerase of the virus. The study identified Methyl Rosmarinate, Cianidanol, Royleanone, and 6,7-Dehydroroyleanone as dual-target inhibitors against any two of the key proteins. At the same time, 7-alpha-Acetoxyroyleanone from Rosmarinus officinalis was found to be a multi-target protein inhibitor against all three proteins. The potential multi-target inhibitor was subjected to molecular dynamics simulations to assess the stability of the protein–ligand complexes along with the corresponding reference ligands. The findings specified stable interactions of 7-alpha-Acetoxyroyleanone with the protein targets. The results based on the in silico study indicate that the phytocompounds can potentially inhibit the essential proteins of the Infectious Bronchitis virus; however, in vitro and in vivo studies are required for validation. Nevertheless, this study is a significant step in exploring the use of botanicals in feed to control Infectious Bronchitis infections in poultry.


Introduction
Infectious Bronchitis (IB) is a highly contagious disease with significant economic implications for the global chicken industry. First documented in 1931, IB is mainly linked to respiratory, reproductive, digestive, and renal disorders in domestic chickens and various avian species [1,2]. Infectious bronchitis virus (IBV) replicates primarily in the epithelial cells of the respiratory tract, resulting in respiratory problems [3]. The epithelial cells in the oviduct and the kidney are also susceptible to IBV infection, impairing the quality and production of eggs and causing nephritis. IB infections can cause a mortality rate of 20-30% [4,5], increasing significantly with secondary infections in infected flocks [6,7]. The IBV is an enveloped positive-sense single-stranded RNA virus (+ssRNA) that belongs to the genus Gammacoronavirus of the Coronaviridae family [8]. The IBV Viruses 2023, 15, 847 3 of 17 and exploration of antiviral drugs. The Mpro, PLpro and the RdRp of IBV are the three active viral proteins that function in the replication, transcription and assembly of the virus. Therefore, these proteins have the potential to be targeted in the development of effective antivirals. Mpro and PLpro are two viral proteases recognized for their ability to carry out the proteolytic maturation and cleavage of polyprotein 1a and 1ab into various NSPs, each of which plays a unique and vital role [10,36,37]. Given the essential role of both proteases, they become an attractive target for developing antivirals against coronaviruses [38][39][40]. In contrast, RdRp is a crucial enzyme required for viral replication and has been investigated as a potential therapeutic target for a number of viruses, including SARS-CoV-2 [41][42][43].
The objective of this study was to use a bioinformatics approach to evaluate the antiviral properties of fourteen botanicals known to enhance the growth rate, production, feed efficiency, health status, and immunological responses of poultry. Several of these botanicals, including common yarrow, garlic, turmeric, and borage, have a history of being characterized as antimicrobial in poultry [44][45][46][47][48][49][50][51][52][53]. In addition, studies have shown that peppermint and chicory can improve the growth performance of broiler chickens without negatively affecting their intestinal morphology. Chicory forage has been found to be a potentially beneficial feed ingredient for broiler chickens [54,55]. Peppermint promotes growth in young broilers, improves their performance and carcass features, and reduces mortality [54,56]. A mixture of oregano, rosemary, and volatile fennel oils has been shown to have potent antibacterial activity against coliform bacteria, improving feed efficiency and carcass quality in broiler chickens [54,[56][57][58][59][60][61][62][63][64]. Additionally, lemon balm have been found to have various impacts on body weight and bone health in broiler chickens [59]. The immunomodulatory properties of black cumin have been demonstrated in broiler chickens, significantly affecting weight gain and FCR [60][61][62]. Garlic powder and tulsi leaf powder were observed to potentially serve as an alternative to conventional antimicrobials, improving the production efficiency and immune status of birds by enhancing the T-cell mediated immune response [47][48][49][50]. It has been demonstrated that sumac (Rhus coriaria) has the potential as a feed additive for broilers, benefiting those raised in stressful situations by improving their growth and feed efficiency [65,66]. Adding marjoram leaf powder to the broiler diet significantly improved FCR, productivity, and daily body weight growth without affecting immune response [46].
As a first step for the study, we screened the selected botanicals based on ADMETpredicted pharmacokinetics and toxicity, which is followed by the template-based molecular docking of phytocompounds of the selected botanicals against Mpro, PLpro, and RdRp from IBV. The findings were further evaluated using molecular dynamics simulation analysis. In order to generate a comparative study, target viral proteins with native ligand inhibitors were selected as reference complexes.

Materials and Methods
This study performed molecular dynamics simulation using the Schrödinger software package version 2022-1 [67] and primary docking using Cresset Flare version 6. A SWISS-MODEL was used for homology modelling to generate a 3D model [68]. The PyMOL (https://pymol.org/, accessed on 25 January 2023) and LigPlot+ were used to visualize and generate binding poses [69].

Ligand Preparation
Botanical plants were collected through a literature survey, and their compounds were sourced from a curated database IMMPAT [70]. A local compound library named PhytoChemiome Library Version 1 (PCLibVer1) was compiled of about 1300 botanical compound ligands, which were retrieved in SDF format from the PubChem database (https://pubchem.ncbi.nlm.nih.gov/, accessed on 13 January 2023) [71]. Cresset Flare software was used for the preparation and docking of the ligands [72][73][74].

Protein Preparation
The 3-dimensional protein structures of IBV Mpro (PDB ID: 2Q6F) and PLpro (PDB ID: 4X2Z) were retrieved from RCSB Protein Data Bank database (https://www.rcsb.org/, accessed on 13 January 2023) [15,75,76]. The structure of Mpro was in complex with a Michael acceptor inhibitor (N3), which was removed from the structure. Since the crystal structure of IBV-RdRp was unavailable, protein homology modeling was performed using SWISS-MODEL based on the SARS-CoV2 RdRp structure (PDB ID: 7DFG) with 63.06% sequence identity [68]. The sequence of IBV-RdRp was obtained from NCBI reference sequence accession no. Np_066134.1. The homology model was subjected to structural refinement using the GalaxyRefine server (http://galaxy.seoklab.org/cgi-bin/submit.cgi?type= REFINE, accessed on 20 January 2023) [77]. The refined tertiary structure was analyzed by the PDBSum (http://www.ebi.ac.uk/thornton-srv/databases/pdbsum/, accessed on 20 January 2023) online server [78,79] ( Figure S1). Subsequently, the water molecules and heteroatoms from all the structures were removed, and the proteins were subjected to energy minimization.

Determining the Target Sites
The binding sites of Michael acceptor inhibitor N3 was determined from the crystal structure of Mpro (2Q6F) using the inbuild utilities of Flare [15]. Myricetin, a known inhibitor of PLpro, was docked into the catalytic triad active site of PLpro [80,81]. The interacting residues were identified and designated as a 'myricetin binding site'. The AT-9010 binding site of SARS-CoV-2 RdRp (PDB ID: 7ED5) was determined using Flare [82]. The site was then aligned with the IBV RdRp to identify the binding residues. The binding sites of Michael acceptor N3, myricetin and AT-9010 were selected as the target sites, while Michael acceptor inhibitor N3 (PubChem CID: 42627499), Myricetin (PubChem CID: 5281672) and AT-9010 (PubChem CID: 162642756) were taken as reference molecules for Mpro, PLpro and RdRp, respectively.

Pharmacokinetic Assessment
The potential biological properties of PCLibVer1 compounds were investigated using ADMETlab 2.0 (https://admetmesh.scbdd.com/, accessed on 17 January 2023) [83]. The server was used to estimate various properties related to pharmacokinetics and toxicity, such as solubility and permeability based on Lipinski's Rule of Five, potential for druginduced liver injury, and the likelihood of carcinogenicity. These predictions can provide valuable information on the potential biological activity and safety of the phytocompound when used in poultry, which are food animals. The screening criteria for ADMET properties were set to stringent standards: no violations of Lipinski's Rule of Five, negative results for potential hepatotoxicity and carcinogenicity.

Virtual Screening
The screening of the PCLibVer1 compounds targeting IBV Mpro, PLpro and RdRp was performed using Cresset Flare software. The template-based docking of the ligands was performed sequentially using the Virtual screening protocol of Flare followed by Normal Docking, Accurate Docking, Extra Precision Docking (five runs) andExtra Precision Docking (ten runs) Protocol. At each stage, only half of the ligands were selected based on Gibbs free energy (∆G) of the binding poses compared to the reference. A final list of ten ligands was generated against each target protein, wherein the ligands were compared, and a common botanical inhibitor against the target proteins was selected. Using the proteinligand interaction tool of Schrödinger Maestro, post-docking investigations were performed, revealing the sizes and positions of binding sites, hydrogen-bond interactions, hydrophobic interactions, and bonding distances [67]. The common compound was evaluated for inhibitory binding with the target proteins via molecular dynamics simulation. The ligand's best and most energetically optimal conformations were selected for molecular dynamics.

Molecular Dynamics
Molecular dynamics (MD) simulation was carried out with the Schrödinger software package. To begin, the protein-ligand complex was prepared using the "protein preparation wizard" feature in Schrödinger Maestro v2022.1. By employing the OPLS3 force field, the pre-processing steps for the protein preparation, including the addition of hydrogen atoms, the assignment of bond order, the removal of water molecules beyond 5.0 Å, and the establishment of a pH value of 7 were performed. Following the pre-processing, a solvated system according to the TIP3P water model was generated using the Desmond System Builder tool, wherein a cubic simulation box having periodic boundary conditions of 10 × 10 × 10 Å was prepared. Adding Cl or Na+ ions neutralized the system, and 0.15 M salts created an isosmotic state. The MD simulation was run using the Desmond molecular dynamics tool at constant pressure (1.013 bar) and temperature (300 K) for a period of 100 ns, with 1000 frames selected for the trajectory [84]. The post-MD analyses, including root mean square deviation (RMSD), root mean square fluctuations (RMSF), and protein-ligand contacts, were analyzed using the simulated interaction diagram tool.

Results
A comprehensive evaluation of the pharmacological properties of PCLibVer1 phytochemicals was performed using the ADMETLab 2.0. The screening criteria were used to select compounds that met specific standards for their ADMET properties, such as solubility, permeability, and potential toxicity. After the screening, 520 phytochemicals were identified as suitable for further analysis. These compounds were then prepared for molecular docking with targeted proteins using the Cresset Flare software.

Virtual Screening of Ligand Libraries
The screening of a set of phytocompounds of PCLibVer1 was conducted against the protein targets using a molecular docking protocol embedded in Flare software. The docking involves predicting how well the compounds bind to the proteins by estimating the strength of the chemical interactions between them. The docking process was carried out in several stages, starting with a basic docking procedure and increasing precision with each subsequent step. In the first phase, the Quick Docking protocol of Flare was employed, which proceeded with Normal Docking, Accurate Docking, Very Accurate Docking, Extra Precision Docking with five runs, and finally, Extra Precision Docking with ten runs. At each stage, the phytocompounds were ranked based on their ∆G binding score, and the top 50% of the phytocompounds with the lowest (more negative) scores were selected. Consequently, a list of the top ten hits (Table S2) was generated for each target protein, and these hits were compared to identify common inhibitor compounds (Table 1). Based on the comparison, several common hits were identified against the protein targets. Specifically, four phytocompounds were found to inhibit the proteins Mpro and PLpro, three molecules inhibited both Plpro and RdRp, and two molecules inhibited both Mpro and RdRp. Out of all the top phytocompounds, only phytocompound 2751796 was predicted to bind effectively to all three protein targets. In order to assess the effectiveness, the phytocompounds were compared to the reference molecules based on the ∆G score. This score reflects the binding affinity of a ligand to a protein and can help predict the potential inhibitory activity of a compound. The analysis revealed that each ligand had a greater or comparable negative ∆G score compared to the reference molecules, indicating a stronger binding affinity with the target protein.

Common Phytocompound Inhibitor against Mpro and Plpro
The binding of phytocompounds to the IBV Mpro and Plpro proteins revealed four common hits. Of these, phytocompound 6479915 (Methyl rosmarinate) from Mentha piperita had the strongest binding to Mpro with a binding energy of −10.135 kcal/mol. The second most efficient binding was seen for phytocompound 23243692 (7-O-Methylrosmanol), which was followed by 23243694 (Epirosmanol), and 46883407 (Rosmaquinone β) with a binding energy of −8.966 kcal/mol, −8.945 kcal/mol and −8.759 kcal/mol, respectively. All of these phytocompounds are derived from the Rosmarinus officinalis botanical. When evaluating the binding of these phytocompounds to the PLpro protein, 23243692 showed the highest binding energy of −8.864 kcal/mol, which was followed by 23243694, 6479915 and 46883407 with a binding energy of −8.773 kcal/mol, −8.688 kcal/mol and −8.49 kcal/mol, respectively. The analysis of molecular interactions between phytocompounds and protein targets showed that phytocompound 6479915 formed hydrogen bonds with GLY44, HIS161 and GLU164 of Mpro, and PHE151, SER152, ASP153, ASN155 and ALA237 of PLpro (Figure 1a,b). Phytocompounds 23243692, 23243694, and 46883407 sharing similar structures showed only hydrophobic interactions with binding site amino acid residues of Mpro and PLpro, except for 46883407 forming two hydrogen bonds with ASN155 of PLpro ( Figure S2b-d). All the ligands, however, showed non-bonded contacts, including van der Waals interactions and weak electrostatic interactions, that could contribute to the specificity of the protein-ligand interaction. A comparison of the phytocompounds binding to their corresponding target sites revealed common interacting residues for both proteins. These residues include HIS41, GLY44, LYS45, TRP51, CYS143, LEU163, ASP185, GLY 186 and GLU187 for Mpro. Similarly, for PLpro, the common interacting residues identified were PHE151, SER152, ASP153, ASN155, PHE256, ALA237, GLY240, THR260 and ILE290.

Common Phytocompound Inhibitor against Mpro and RdRp
The binding affinities of the two common phytocompounds were evaluated against the protein targets Mpro and RdRp. The findings revealed that the phytocompound 9064 (Cianidanol) exhibited binding energy of −9.497 kcal/mol and −9.389 kcal/mol against Mpro and RdRp, respectively, while the 13820511 (Isorosmanol) phytocompound exhibited binding energy of −9.191 kcal/mol and −8.154 kcal/mol against Mpro and Plpro, respectively. The analysis of molecular interactions between the phytocompounds and the protein targets revealed that the phytocompound 9064 formed hydrogen bonds with amino acid residues ASN26, HIS41, GLY44, GLY141, GLU187 and ASP185 of Mpro as well as with residues ASP41, LYS55, CYS58, THR215 and ASN218 of RdRp (Figure 2a,b). The phytocompound 13820511, on the other hand, formed hydrogen bonds with the amino acid residue HIS41 and LYS45 of the Mpro protein, and it showed only hydrophobic interactions with the binding site amino acid residues of the RdRp protein ( Figure S3b). A comparison of the phytocompounds binding to their corresponding target sites revealed common interacting residues for both proteins. The analysis showed that for Mpro, both phytocompounds interacted with the amino acid residues ASN25, ASN26, HIS41, GLY44, LYS45, CYS143, ASP185, GLU186 and GLU187. Similarly, For RdRp, the common residues were VAL42, ASP41, PHE40, THR215, ASP217 and ASP227 ( Figure S3c).

Common Phytocompound Inhibitor against Mpro and RdRp
The binding affinities of the two common phytocompounds were evaluated against the protein targets Mpro and RdRp. The findings revealed that the phytocompound 9064 (Cianidanol) exhibited binding energy of −9.497 kcal/mol and −9.389 kcal/mol against Mpro and RdRp, respectively, while the 13820511 (Isorosmanol) phytocompound exhibited binding energy of −9.191 kcal/mol and −8.154 kcal/mol against Mpro and Plpro, respectively. The analysis of molecular interactions between the phytocompounds and the protein targets revealed that the phytocompound 9064 formed hydrogen bonds with amino acid residues ASN26, HIS41, GLY44, GLY141, GLU187 and ASP185 of Mpro as well as with residues ASP41, LYS55, CYS58, THR215 and ASN218 of RdRp (Figure 2a,b). The phytocompound 13820511, on the other hand, formed hydrogen bonds with the amino acid residue HIS41 and LYS45 of the Mpro protein, and it showed only hydrophobic interactions with the binding site amino acid residues of the RdRp protein ( Figure S3b). A comparison of the phytocompounds binding to their corresponding target sites revealed common interacting residues for both proteins. The analysis showed that for Mpro, both phytocompounds interacted with the amino acid residues ASN25, ASN26, HIS41, GLY44, LYS45, CYS143, ASP185, GLU186 and GLU187. Similarly, For RdRp, the common residues were VAL42, ASP41, PHE40, THR215, ASP217 and ASP227 ( Figure S3c).

Common Phytocompound Inhibitor against Mpro and RdRp
The binding affinities of the two common phytocompounds were evaluated against the protein targets Mpro and RdRp. The findings revealed that the phytocompound 9064 (Cianidanol) exhibited binding energy of −9.497 kcal/mol and −9.389 kcal/mol against Mpro and RdRp, respectively, while the 13820511 (Isorosmanol) phytocompound exhibited binding energy of −9.191 kcal/mol and −8.154 kcal/mol against Mpro and Plpro, respectively. The analysis of molecular interactions between the phytocompounds and the protein targets revealed that the phytocompound 9064 formed hydrogen bonds with amino acid residues ASN26, HIS41, GLY44, GLY141, GLU187 and ASP185 of Mpro as well as with residues ASP41, LYS55, CYS58, THR215 and ASN218 of RdRp (Figure 2a,b). The phytocompound 13820511, on the other hand, formed hydrogen bonds with the amino acid residue HIS41 and LYS45 of the Mpro protein, and it showed only hydrophobic interactions with the binding site amino acid residues of the RdRp protein ( Figure S3b). A comparison of the phytocompounds binding to their corresponding target sites revealed common interacting residues for both proteins. The analysis showed that for Mpro, both phytocompounds interacted with the amino acid residues ASN25, ASN26, HIS41, GLY44, LYS45, CYS143, ASP185, GLU186 and GLU187. Similarly, For RdRp, the common residues were VAL42, ASP41, PHE40, THR215, ASP217 and ASP227 ( Figure S3c).

Common Phytocompound Inhibitor against Mpro, PLpro and RdRp
The phytocompound 2751796 (7alpha-Acetoxyroyleanone) from the Rosmarinus officinalis botanical exhibited strong binding affinity against all the IBV enzyme targets. The binding affinity of the phytocompound against Mpro was found to be −8.608 kcal/mol, which was comparable to that of the reference inhibitor (−9.698 kcal/mol). Additionally, the phytocompound exhibited a binding affinity of −8.703 kcal/mol and −8.9 kcal/mol against PLpro and RdRp, respectively. These values are higher than those of the cor- These results indicate the multi-target inhibitory potential of the phytocompound 2751796 against these enzymes. Furthermore, the analysis of the molecular interactions revealed the presence of specific interactions between the phytocompound 2751796 and the target proteins. These interactions include hydrogen bonds, salt bridges, pi-cation and hydrophobic interactions. Specifically, the phytocompound formed two hydrogen bonds with the amino acids HIS41 and ASP185, a salt bridge with HIS41, pi-cation interactions with HIS41, LYS45, and hydrophobic interactions with the ASN25, LEU163 and GLU187 amino acid residues of Mpro (Figure 4a). The phytocompound formed four hydrogen bonds with the amino acids ASP153, ASN155, THR238 and GLY240 and exhibited hydrophobic interactions with ASP153, PHE236, THR238, PHE256 and ILE290 of PLpro (Figure 4b). In the case of RdRp, the phytocompound formed four hydrogen bonds with ASP41, CYS43, CYS43 and THR215 and exhibited hydrophobic interactions with VAL42, ASN44, VAL213, and ASP217 (Figure 4c).

Common Phytocompound Inhibitor against Mpro, PLpro and RdRp
The phytocompound 2751796 (7alpha-Acetoxyroyleanone) from the Rosmarinus officinalis botanical exhibited strong binding affinity against all the IBV enzyme targets. The binding affinity of the phytocompound against Mpro was found to be −8.608 kcal/mol, which was comparable to that of the reference inhibitor (−9.698 kcal/mol). Additionally, the phytocompound exhibited a binding affinity of −8.703 kcal/mol and −8.9 kcal/mol against PLpro and RdRp, respectively. These values are higher than those of the corresponding reference compounds, −7.672 kcal/mol and −2.265 kcal/mol, respectively. These results indicate the multi-target inhibitory potential of the phytocompound 2751796 against these enzymes. Furthermore, the analysis of the molecular interactions revealed the presence of specific interactions between the phytocompound 2751796 and the target proteins. These interactions include hydrogen bonds, salt bridges, pi-cation and hydrophobic interactions. Specifically, the phytocompound formed two hydrogen bonds with the amino acids HIS41 and ASP185, a salt bridge with HIS41, pi-cation interactions with HIS41, LYS45, and hydrophobic interactions with the ASN25, LEU163 and GLU187 amino acid residues of Mpro (Figure 4a). The phytocompound formed four hydrogen bonds with the amino acids ASP153, ASN155, THR238 and GLY240 and exhibited hydrophobic interactions with ASP153, PHE236, THR238, PHE256 and ILE290 of PLpro (Figure 4b). In the case of RdRp, the phytocompound formed four hydrogen bonds with ASP41, CYS43, CYS43 and THR215 and exhibited hydrophobic interactions with VAL42, ASN44, VAL213, and ASP217 (Figure 4c).

Molecular Dynamics Analysis
The root mean square deviation (RMSD) values of the Mpro-2751796, PLpro-2751796, and RdRp-2751796 complexes were evaluated throughout the 100 ns molecular dynamics (MD) simulation. These RMSD values were used to determine the equilibration of the protein C-alpha atoms for each complex compared to their respective reference complexes. The findings revealed that the equilibration of the protein C-alpha atoms for Mpro-2751796 occurred at an RMSD value of 1.99 ± 0.27 Å, which was lower than the Mpro-N3 reference complex (2.39 ± 0.39 Å) (Figure 5a). Likewise, the equilibration of the protein C-

Molecular Dynamics Analysis
The root mean square deviation (RMSD) values of the Mpro-2751796, PLpro-2751796, and RdRp-2751796 complexes were evaluated throughout the 100 ns molecular dynamics (MD) simulation. These RMSD values were used to determine the equilibration of the protein C-alpha atoms for each complex compared to their respective reference complexes. The findings revealed that the equilibration of the protein C-alpha atoms for Mpro-2751796 occurred at an RMSD value of 1.99 ± 0.27 Å, which was lower than the Mpro-N3 reference complex (2.39 ± 0.39 Å) (Figure 5a). Likewise, the equilibration of the protein C-alpha atoms for PLpro-2751796 occurred at an RMSD value of 2.48 ± 0.51 Å, which was higher than the PLpro-myricetin reference complex (2.14 ± 0.33 Å). The phytocompound displayed comparable binding to the PLpro up to 60ns but showed higher fluctuations from 60 to 100 ns (Figure 5b). The equilibration of the protein C-alpha atoms for RdRp-2751796 occurred at an RMSD value of 3.29 ± 0.33 Å, which was slightly higher than the RdRp-AT9010 reference complex (3.04 ± 0.31 Å) (Figure 5c). These results provide valuable information about the structural stability of these complexes and suggest that the 2751796 phytocompound has a relatively stable residence in the binding pocket of the target proteins at the given docking pose.
displayed comparable binding to the PLpro up to 60ns but showed higher fluctuations from 60 to 100 ns (Figure 5b). The equilibration of the protein C-alpha atoms for RdRp-2751796 occurred at an RMSD value of 3.29 ± 0.33 Å, which was slightly higher than the RdRp-AT9010 reference complex (3.04 ± 0.31 Å) (Figure 5c). These results provide valuable information about the structural stability of these complexes and suggest that the 2751796 phytocompound has a relatively stable residence in the binding pocket of the target proteins at the given docking pose. Upon examining the residue-based root mean square fluctuation (RMSF) for each investigated complex, no significant differences were observed. However, the binding of the 2751796 phytocompound to Mpro caused a slight loss of mobility (Figure 6a). In contrast, binding the phytocompound to PLpro and RdRp did not cause any significant change in the flexibility of the amino acid residues compared to the binding of the reference compounds (Figure 6b-c). These results are consistent with the analysis of secondary structure element (SSE) distribution, which showed no significant change in the secondary structure of the protein upon binding of the phytocompound compared to the reference compounds.
During the 100 ns MD simulation, hydrogen bonds, hydrophobic interactions including pi-cation, pi-pi stacking, water bridges and ionic interactions formed by the ligand with the amino acids of the proteins were analyzed. The phytocompound bound comparably to the reference compounds in the case of Mpro, as indicated by the number of amino acid contacts and the retention of the interaction throughout the simulation time (measured as interaction fractions) (Figure 7a). On the other hand, in the case of PLpro and RdRp, the phytocompound interacted with a greater number of amino acid residues than the reference compounds, indicating a stronger binding of the ligand to those amino acids (Figure 7b,c). Although the reference compounds interacted with fewer residues, the binding was observed over a larger portion of the trajectory, indicating consistent binding. Notably, ionic interactions were absent in the protein-reference and protein-phytocompound complexes among the variety of interactions analyzed. Upon examining the residue-based root mean square fluctuation (RMSF) for each investigated complex, no significant differences were observed. However, the binding of the 2751796 phytocompound to Mpro caused a slight loss of mobility (Figure 6a). In contrast, binding the phytocompound to PLpro and RdRp did not cause any significant change in the flexibility of the amino acid residues compared to the binding of the reference compounds (Figure 6b,c). These results are consistent with the analysis of secondary structure element (SSE) distribution, which showed no significant change in the secondary structure of the protein upon binding of the phytocompound compared to the reference compounds.  During the 100 ns MD simulation, hydrogen bonds, hydrophobic interactions including pi-cation, pi-pi stacking, water bridges and ionic interactions formed by the ligand with the amino acids of the proteins were analyzed. The phytocompound bound comparably to the reference compounds in the case of Mpro, as indicated by the number of amino acid contacts and the retention of the interaction throughout the simulation time (measured as interaction fractions) (Figure 7a). On the other hand, in the case of PLpro and RdRp, the phytocompound interacted with a greater number of amino acid residues than the reference compounds, indicating a stronger binding of the ligand to those amino acids (Figure 7b,c). Although the reference compounds interacted with fewer residues, the binding was observed over a larger portion of the trajectory, indicating consistent binding. Notably, ionic interactions were absent in the protein-reference and protein-phytocompound complexes among the variety of interactions analyzed.

Discussion
Among poultry diseases, IB is a widespread disease affecting poultry globally, causing high morbidity rates and mortality along with decreased egg and meat production. Currently, the control of IBV is essentially attempted using live attenuated and inactivated vaccines. However, the efficacy of these vaccines is being limited by the increasing genetic diversity of IBV and the emergence of new IBV variants. In view of these difficulties, it is crucial to find new ways of controlling IB in chickens. This study aimed to investigate the antiviral effect of phytocompounds from fourteen botanicals against the replication of the IBV in poultry. These botanicals were selected based on previous research indicating their potential to improve poultry growth rate, feed-to-gain ratio, health status, and immune system function (Table S1). Initially, the study assessed the pharmacological properties of the phytochemicals and identified 520 phytocompounds that met specific standards for their ADMET properties. Molecular docking analysis revealed several common hits against IBV Mpro, PLpro and RdRp. Specifically, phytocompounds 6479915, 23243692, 23243694, and 46883407 were found to inhibit both Mpro and PLpro, while phytocompounds 9064 and 13820511 exhibited inhibition against Mpro and RdRp. Additionally, phytocompounds 442084, 2751794, and 75552 were identified as common inhibitors against the viral PLpro and RdRp enzyme targets. Of all the potential inhibitors, only Phytocompound 2751796 was predicted to bind effectively to all three protein targets.
The prediction of ADMET properties, such as solubility, permeability, liver injury potential, and carcinogenicity, plays a critical role in evaluating the biological activity and safety of phytocompounds. It is essential to consider these predictions to make informed decisions regarding the use of phytocompounds in poultry production and to ensure the health and well-being of these animals. The pharmacokinetic properties including the oral absorption, solubility, permeability and systemic distribution potential of the phytocompounds in poultry were determined by using Lipinski's Rule of Five. This rule states that drugs with desirable pharmacokinetic properties typically have a molecular weight of less than 500 Da, a logP value (the octanol-water partition coefficient) of less than 5, no more than 10 hydrogen bond donors and no more than 5 hydrogen bond acceptors [85]. Additionally, prediction of potential toxicity was estimated to evaluate the likelihood of drug-induced liver injury and carcinogenicity. Drug-induced liver injury is a common adverse effect of many drugs and can have severe consequences for animal health. Carcinogenicity, on the other hand, is the potential of a substance to cause cancer and is an essential consideration for substances used in food animals.
The molecular docking study evaluated the binding energy and molecular interactions between the phytocompounds and target proteins and identified potential inhibitors. Among the evaluated inhibitors, Methyl Rosmarinate (PubChem CID: 6479915) was found to be an effective inhibitor of both Mpro and PLpro. Methyl Rosmarinate is a phytocompound derived from the Mentha piperita plant, which is commonly known as peppermint. It is a phenolic compound with antioxidant properties and has been studied for its potential health benefits. Methyl rosmarinate shows antioxidative and antifungal activities. It has inhibitory activities against tyrosinase, α-glucosidase, and matrix metalloproteinase-1 (MMP-1) [86][87][88][89][90]. The phytocompound Cianidanol (PubChem CID 9064) demonstrated potency as an inhibitor of Mpro and RdRp. Cianidanol is an antioxidant flavonoid in the Ocimum sanctum plant, which is also known as Holy basil or Tulsi. This plant is widely used in traditional medicine, particularly in India and other parts of Asia, due to its medicinal properties. Studies have shown that cianidanol has the ability to inhibit both the Mpro and spike proteins of SARS-CoV-2 as well as play a therapeutic and immunomodulatory role in chronic hepatitis [91][92][93]. At the same time, Royleanone (PubChem CID: 442084) and 6,7-Dehydroroyleanone (PubChem CID: 2751794) were identified as potent inhibitors of Plpro and RdRp. Another royleanone phytocompound, 7-alpha-Acetoxyroyleanone (PubChem CID: 2751796), was identified with multi-target inhibitory potential against all three IBV target proteins viz Mpro, Plpro and RdRp. The royleanones are a diterpenoids class of compounds found in the Rosmarinus officinalis plant, which is commonly known as rosemary. Studies have shown that these compounds have the potential to inhibit P-Glycoprotein, which may combat multidrug resistance (MDR) and induce cytotoxic effects in cancers [94,95].
The MD simulation studies demonstrate that 7-alpha-Acetoxyroyleanone exhibits stable binding to the binding pockets of the target protein as compared to known inhibitor reference compounds. Specifically, the RMSD analysis indicates the steady binding of 7-alpha-Acetoxyroyleanone to Mpro and RdRp throughout the MD simulation, and to Plpro for most of the simulation. The RMSF analysis characterizes protein backbone mobility and indicates the minimal effect of 7-alpha-Acetoxyroyleanone binding on the flexibility of target proteins, which was consistent with the secondary structure analysis. The findings also reveal various interactions, such as hydrogen bond interactions and non-bonded contacts, providing stability to the binding of 7-alpha-Acetoxyroyleanone to the target protein. These results validate that 7-alpha-Acetoxyroyleanone may serve as a multi-target inhibitor against Mpro, Plpro, and RdRp of IBV.
The research findings suggest that cianidanol, methyl rosmarinate, and royleanones hold promise as nutraceuticals for therapeutic purposes against the infectious bronchitis virus, potentially serving as a natural substitute for antibiotics. Additionally, the botanicals Ocimum sanctum, Mentha piperita, and Rosmarinus officinalis can be used as feed additives as a prophylactic strategy against the virus. In summary, using natural phytocompounds and botanicals in poultry farming shows great potential to combat IBV infection while ensuring the health and welfare of the poultry.

Supplementary Materials:
The following supporting information can be downloaded at: https:// www.mdpi.com/article/10.3390/v15040847/s1, Table S1: List of Botanicals included in the study: Table S2: List of top 10 hits against each target protein of IBV. The ligands are identified by their Pubchem Compound Identifier (CID); Figure S1: Analysis of IBV RdRp tertiary structure using PDBsum. (a) Secondary structure diagram of the protein, with pink arrows representing strands and purple springs representing helices, and various motifs depicted in red. (b) Assessment of refined protein structure through Ramachandran plot analysis, with core allowed regions in red, disallowed regions in brown, and regions with limited utilization in pale yellow; Figure

Data Availability Statement:
The data presented in this study are openly available in the National Center for Biotechnology Information (NCBI) and Indian Medicinal Plants, Phytochemistry and its Therapeutics (IMPPAT).